A taxonomic reevaluation of the genus Ostrya (Betulaceae) in the Southwestern

United States based on leaf morphology

Stephanie Dykema and Ross A. McCauley

Fort Lewis College, Department of Biology, Durango, CO

RESULTS DISCUSSION

LITERATURE CITED

ACKNOWLEDGEMENTS

Funding for this project was provided by the Fort Lewis College Faculty

Development Grant program. Research permits were granted by Grand Canyon

N.P., Guadalupe Mountains N.P., and Big Bend N.P. to R. McCauley.

BACKGROUND AND INTRODUCTION

METHODS

Ostrya is a genus of small deciduous trees in the birch family, Betulaceae, commonly

known as hop-hornbeam. The genus represents an ancient Holarctic plant distribution

and occurs in deciduous forests of Asia and Europe as well as in eastern North

America in abundance with isolated populations found in the dryer mountains of the

southwestern United States and NW Mexico, extending south as a cloud forest

endemic through central Mexico to Honduras (Furlow, 1990). The number of species

within North American Ostrya is small (3-5) and has been debated by systematists.

Ostrya populations in the eastern U.S. have been commonly recognized as Ostrya

virginiana (Mill.) K. Koch (Brain and Spamer, 2000) however, the grouping of

populations in the southwest has not been so easily agreed upon.

The isolated and disjunct occurrences of Ostrya across the SW US have led to the

recognition of one to three distinct species (Figure 1 & 2). Some botanists group the

southwestern populations under the single species Ostrya knowltonii Coville. The

genus has been further segregated, however, by some researchers based primarily on

geographic location and variation in fine-scale leaf morphology. Populations of the

Guadalupe Mountains have been referred to as O. baileyi Rose and populations

restricted to the Chisos Mountains in Texas and the Sierra del Carmen in Coahuila as

O. chisosensis. Under this segregated view O. knowltonii is restricted to the Colorado

River drainage in northern Arizona and southern Utah. No systematic study has been

conducted to consider the distinction of each population in relation to each other and

evaluate the validity of the proposed taxonomic delineation (Furlow, 1990).

Figure 2. Species of Ostrya illustrating habit, leaf morphology, and

reproductive structures. A. Collecting O. knowltonii near Moab, UT B.

Infructescence O. knowltonii, Alamo Canyon, NM C. O. baileyi with

staminate catkin at type locality in Dog Canyon, Guadalupe Mountains

National Park, TX D. O. knowltonii in Roaring Springs Canyon, Grand

Canyon National Park, AZ.

A.

Morphometric analyses of Ostrya in the SW US provides evidence that one variable

species likely exists in this region (Figure 3-6). Though several taxa have been

proposed for the SW US our data indicates that insignificant morphological variation

exists among populations across the region to support the segregation of species.

The distinction of O. baileyi in the Guadalupe Mountains and O. chisosensis in the

Chisos Mountains was based principally off of differences in the expression of stipitate

glands on the twigs and petioles with O. baileyi showing prominent glands and O.

chisosensis lacking these glands (Rose, 1905; Correl 1965). Additional variations in

leaf shape, leaf margin characteristics, and pubescence were also suggested (Furlow

1990) however, these characteristics were merely observed in few collected

specimens. Our field observations and morphological analysis indicated that these

characters are in fact variable within populations and thus do not serve as a means for

segregating taxon groups.

Phenotypic plasticity is likely responsible for much of the morphological variation

observed across the sampled populations. One particular example was the consistent

divergence of the Oak Creek Canyon, AZ population from the remaining O. knowltonii

populations. The ecosystem where this population grows is more moist and highly

vegetated than the open environments more commonly inhabited by Ostrya in the SW

(Figure 7). The larger trees and larger leaves are likely a physiological response to

lower light levels in this ecosystem which approaches the understory habitat of O.

virginiana and represents an inherent adaptability within the genome of the species.

This study supports the view that the SW US populations of Ostrya represent a single

taxon which should be referred to as Ostrya knowltonii Coville. Additional molecular

phylogenetic data for these populations inferred from four independent gene regions

additionally showed only minor nonspecific variation across the region (Bielz and

McCauley, 2013). Likely these isolated and disjunct occurrences represent relictual

habitat islands representing the remnants of a wider continuous distribution contracted

by long-term climatic changes.

Future investigations of morphological variation in North American Ostrya are needed

to test the hypothesis that isolated occurrences of Ostrya in the Madrean Sky Islands

and Copper Canyon region of NW Mexico represent a southern extension of O.

knowltonii. Additionally the circumscription and affinities of the cloud forest

occurrences across the Sierra Madre Occidental, Sierra Madre Oriental, and Southern

Cordillera of Mexico currently viewed as conspecific with O. virginiana need to be

evaluated. Figure 1. Distribution of Ostrya in

the SW United States and NW Mexico

based on herbarium records.

A B

A B

C

D

E

F

A.

Figure 6. Results of univariate morphometric analyses using 11 measured variables and 6 composite ratio

variables on leaf morphology in 10 populations of Ostrya. Shaded areas indicate the mean and 25th and

75th percentiles of the data with whiskers representing the 10th and 90th percentiles. Letters associated with

boxes indicate groupings of similar values determined through a Tukey Post Hoc Test (P<0.001). Several

characteristics show variation only between the Eastern populations and the Southwestern populations

however the WV and OH populations are significantly similar to most of the SW populations for the majority

of the measured characteristics. Variation between the Eastern O. virginiana and the Grand Canyon

populations as well as the Utah population can be seen for many of the characteristics. The Oak Creek

Canyon, AZ population is significantly different from the rest of the SW populations but similar to the

Eastern populations for petiole length and length of largest tooth and is significantly different from all other

populations for length of smallest tooth.

Figure 3. Multivariate analysis of ratio data. PCA. Centered data.

Log10 transformed. No pattern is apparent with the SW populations but

the O. virginiana of WV and OH show distinct segregation.

Figure 4. Multivariate analysis of continuous data. PCA. Centered data.

Log10 transformed. O. virginiana of the Eastern states shows clear

segregation. O. knowltonii of Oak Creek Canyon, AZ trends toward variation

of O. virginiana segregating from the remaining SW populations.

Figure 5. Multivariate analysis of continuous and discrete data. PCoA. O.

virginiana of OH and WV is distinctly segregated. The Oak Creek Canyon, AZ

population shows variation but is still united with the rest of the SW populations.

A B

C

D

BIELZ, A. and R. A. MCCAULEY. 2013. Using DNA barcoding to evaluate species circumscription in

the genus Ostrya (Betulaceae) in the southwestern US. Fort Lewis College Department of Biology

Senior Research Poster.

BRIAN, N.J. AND E.E. SPAMER. 2000. Knowlton Hop-Hornbeam Revisited (Ostrya

knowltonii Cov.). Bartonia 60: 49-56.

CORREL, D.S. 1965. Some additions and corrections to the flora of Texas. Wrightia 3: 126-129.

FURLOW, J. J. 1990. The genera of Betulaceae in the southeastern United States. Journal of the

Arnold Arboretum 71: 1-67.

ROSE, J. N. 1905. Notes on Ostrya, with two new species. Contributions from the United States

National Herbarium 8: 291-293.

• Ten populations representing all known areas of occurrence and each putative

taxon were sampled in the field during the summers of 2012 and 2013. Ten

branches, each from a separate tree, were collected from each population and were

pressed and dried for preservervation in the Fort Lewis College Herbarium (FLD).

• Morphological measurements were recorded for the largest leaf of each sample.

The following morphological characteristics were measured: petiole length (mm),

leaf lamina length (mm), leaf lamina width at widest point (mm), lamina width +30°

(mm), lamina width +60° (mm), lamina width -30° (mm), lamina width -60° (mm),

number of primary leaf veins (one side of leaf), stipitate glands on petiole (no=0,

yes=1), stipitate glands on leaf lamina (no=0, yes=1), abaxial pubescence desity

(lite=0, moderate=1, dense=2), adaxial pubescence density (lite=0, moderate=1,

dense=2), number of margin teeth (between the 3rd and 4th secondary veins on

one side of leaf), depth of largest measured tooth (mm), depth of smallest

measured tooth (mm). To eliminate gross size differences ratios were calculated

between total lamina length and width and lamina length and width at +30°, +60°, -

30°, and -60°.

• Univariate analyses were conducted by constructing box plots representing

variation within individual populations using SigmaPlot Ver. 12. Differences among

the means of populations were determined using a One-way Analysis of Variance

followed by a Tukey Post hoc test to determine significant differences using SPSS

Ver. 18.

• Three multivariate analyses were conducted using MVSP (MultiVariate Statistical

Package) Ver. 3.22: 1) Principal Components Analysis (PCA) of all continuous data.

Data was centered and log10 transformed. 2) PCA of ratio data. Data was centered

and log10 transformed and 3) Principal Coordinates Analysis (PCoA) including all

continuous and discrete data (minus ratios) using the Gower General Similarity

Coefficient.

Figure 7. Ostrya

habitat in Oak

Creek Canyon, AZ

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Le

af

Le

ng

th (

mm

)

20

40

60

80

100

120

140

160

A. Leaf Length a a

b b

c

b b

b b

b

c

c

c

c c c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Wid

th o

f L

ea

f (m

m)

10

20

30

40

50

60

70

80

B. Leaf Width a

a

b b

b b

b

b

b

c

c c c

c c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Wid

th o

f L

ea

f a

t +

30

° (m

m)

5

10

15

20

25

30

35

40

C. Width +30°

a

a

b b

b b

b

b

b

c

c

c

c c

c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Wid

th o

f L

ea

f a

t +

60

°

0

10

20

30

40

50

60

D. Width +60° a a

b b b b

b

b

b

b

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Wid

th o

f L

ea

f a

t -3

0°

5

10

15

20

25

30

35

40

E. Width -30° a

a

b b

b

b b

b

b

b

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Wid

th o

f L

ea

f a

t -6

0°

(mm

)

0

10

20

30

40

50

F. Width -60° a

a b b

b

b b

b

b

b

c c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Pe

tio

le L

eng

th (

mm

)

0

2

4

6

8

10

12

14

G. Petiole Length a

b

a

a a a

a

b

c

c c

c

c c

c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Nu

mb

er

of

Ve

ins o

n O

ne

Sid

e o

f L

ea

f

6

8

10

12

14

16

18

20

22

H. Number of Veins a a

b b

b

b b

b b

c

c c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

ynNum

be

r o

f T

ee

th B

etw

ee

n the

3rd

and

4th

Pri

ma

ry V

ein

s

2

4

6

8

10

12

14

16

I. Number of Teeth

a

a a

a a a

a

a a

b

b

b b

b

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Le

ng

th o

f S

ma

llest T

oo

th (

mm

)

0

1

2

3

4

5

J. Length of Smallest

Tooth

a a

a a

a

a a

a

b b b

b b

b b b

c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Le

ng

th o

f L

arg

est T

oo

th (

mm

)

1

2

3

4

5

6

K. Length of Largest

Tooth a

a

a a

a

a a

b b

b

b

b b

b

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Ra

tio

of L

ea

f L

eng

th to

Le

af W

idth

100

120

140

160

180

200

220

240

260

280

L. Length to Width

Ratio a

a a

b

a

b

a

c c

c c d d

d

d d d

d

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UMO D

og C

yn

O. b

aley

i - G

UMO P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Ra

tio

of L

ea

f W

idth

at +

30

° a

nd

Le

af L

eng

th

20

22

24

26

28

30

32

34

36

38

40

42

M. Ratio of Leaf

Width +30°

a a

a a

a

b

b

c

c

c

c

c c

c

c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UMO D

og C

yn

O. b

aley

i - G

UMO P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Ra

tio

of L

ea

f W

idth

at -3

0°

and

Le

af L

eng

th

20

25

30

35

40

45

N. Ratio of Leaf

Width -30°

a a a

a

a

a a

b

b

b b

b b

b

c

c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UMO D

og C

yn

O. b

aley

i - G

UMO P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Ra

tio

of L

ea

f W

idth

at +

60

° a

nd

Le

af L

eng

th

25

30

35

40

45

50

O. Ratio of Leaf

Width +60°

a a

a

a a

b

b c

c

c

c

c

c c c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UMO D

og C

yn

O. b

aley

i - G

UMO P

ine

Springs

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Ra

tio

of L

ea

f W

idth

at -6

0°

and

Le

af L

eng

th

20

25

30

35

40

45

50

55

60

P. Ratio of Leaf

Width -60°

a a a

a a

b

b c c

c c c c

c

O. v

irginiana

- W

V

O. v

irginiana

- OH

O. b

aley

i - G

UM

O D

og C

yn

O. b

aley

i - G

UM

O P

ine

Spr

ings

Cyn

O. c

hiso

sens

is

O. k

nolw

tonii -

Alam

o Cyn

O. k

nowlto

nii -

UT

O. k

nowlto

nii -

GRCA B

right

Ang

el

O. k

nowlto

nii -

GRCA R

oarin

g Spr

ings

O. k

nowlto

nii -

Oak

Cre

ek C

yn

Ra

tio

of N

um

be

r o

f T

ee

th B

etw

ee

n 3

rd a

nd

4th

Pri

ma

ry V

ein

s a

nd

Le

af L

eng

th

0

5

10

15

20

25

30

Q. Number of Teeth

Ratio

a a

a a

a

a

a

a

b

b

b

b

b b

b b

The purpose of this study was to

examine the distribution and

morphological variation of Ostrya in the

southwestern United States to gain a

clear understanding of the taxonomic

delineation of the species. This work

represents a contribution to an ongoing

systematic revision of the genus across

North America and consisted of

morphometric analyses of leaf form

across ten populations of Ostrya.