Journal of the Oklahoma Native Plant Society, Volume 5, Number 1, December 2005


Oklahoma Native Plant Record  39
Volume 5, Number 1, December 2005 

Murray, C.L. 

https://doi.org/10.22488/okstate.17.100038

A VEGETATION ANALYSIS OF A PIMPLED PRAIRIE 

IN NORTHEASTERN OKLAHOMA 
A thesis for the Department of Life Sciences 

The University of Tulsa 1974 

Constance Lucile Murray 
Current address Tulsa Community College 

The effect of pimple mound microrelief on the vegetation 

of a tall grass prairie was considered. Taxonomic analysis of 

the vegetation affirmed the observation that mound and 

intermounds support communities with differing species 

composition. The difference in the percent cover by living 

vegetation on mounds and intermounds was determined not to be 

statistically significant. The physical composition of the 

soil in the two regions was found to be similar. Two factors 

are suggested as influencing the differences in mound and 

intermound vegetation: that mound soils can provide more 

available water to plants than can intermound soils, and that 

mounds, but not intermounds, contain the burrows of small 

mammals and are modified by their presence. 

INTRODUCTION 

Pimple Mounds are low, 

regular domes of soil that 

establish a microrelief 

pattern common on prairies 

in eastern Oklahoma (Figs. 

1, 2). Similar mounds occur 

throughout the United States 

(Fig. 3), on both level and 

gently sloping terrain 

(Knechtel 1952). Mounds have 

also been described from 

South America (Scheffer 

1958) and Australia 

(Prescott, 1931). They are 

variously called natural 

mounds and hog wallow relief 

in California (Branner, 

1905), Mima mounds in 

Washington (Dalquest and 

Scheffer, 1942), and pimpled 

prairies in the South 

including Oklahoma 

(Knechtel, 1952; Barclay, 

1938). The pimpled prairies 

of Oklahoma are no older 

than late Pleistocene 

(Knechtel, 1952). 

Mounds have been 

examined with basal 

diameters ranging from 10 

feet (3.05 m) to 130 feet 

(39.6 m) (Ross et al., 1968) 

and heights ranging from two 

feet (0.61 m) to seven feet 

(2.13 m) (Dalquest and 

Scheffer, 1942). The 

dimensions of eastern 

Oklahoma mounds correspond 

to the lower figures of 

these ranges (Knechtel, 

1952).



40                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

 
 

 

 Figure 1   Study site showing pimpled microrelief, April 1972 

 

 
 

   Figure 2   Pimple mound, demonstrating greener vegetation than the surrounding 

   prairie, late June 1972



Oklahoma Native Plant Record          41   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

 

 Pimple mounds typically 

consist of unstratified 

soil, variously 

characterized as “loess-

like” (Knechtel, 1952) or 

extremely loose and friable 

(Ross et al., 1968); this 

dark material is responsible 

for most of the height of 

the mound and often rests on 

a clay pan which is lighter 

in color (fig. 4). Between 

the mounds is an intermound 

furrow system. The interface 

between the soil and the 

subsoil does not demonstrate 

the pimpled microrelief; 

thus the mounds are features 

of the soil and not of the 

underlying strata; the soil 

of mound and intermound 

areas is similar in texture 

and composition (Melton, 

1954).  

 Reference in scientific 

literature to natural mounds 

 

 

 

 

 

 
 

 

Figure 3 Shaded areas indicate the major regions in which pimple mounds occur in 

the United States (after Fenneman, 1931) 

 

 

 



42                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

Figure 4 Stylized profile of a mound showing general shape and 

distribution of soil types. Note depression in mound center, probably due 

to disturbance by small mammals. Composite sketch from various authors 

(Scale 1:90)



Oklahoma Native Plant Record          43   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

first appeared in 1865 in 

the California Journal of 

Geology (Melton, 1954). 

Since that time 

controversies concerning the 

origin of these mounds have 

erupted several times, as 

evidenced by the popularity 

of the topic in the 

literature. The contenders 

in the controversies were 

usually geologists, 

variously fueled by 

anthropologists, botanists 

and zoologists. Over the 

years the origins proposed 

for mounds have ranged 

widely, each proposal with 

its staunch defenders. That 

fact that similar mounds 

occur in regions which vary 

greatly in geologic history, 

present geologic forces, 

climate, and flora and fauna 

has made it difficult to 

ascribe one origin to all 

mounds, as most authors have 

attempted to do. A 

discussion of the more 

prominent scientific 

hypotheses advanced for the 

origin of pimpled mounds is 

in order.  

 Campbell (1906)reviewed 

the suggestion that the 

pimple mounds were Indian 

burial mounds; the artifacts 

which usually characterize 

such sites have been found 

in a few mounds excavated in 

the South. However, 

artifacts are absent in most 

of the mounds throughout the 

country (Melton, 1954).  

 Veatch (1906) proposed 

that ants or termites 

construct pimple mounds. The 

town ant, Atta texana, may 

create soil mounds in the 

south (Cain, 1974). These 

mounds are smaller in height 

and larger in diameter than 

typical pimple mounds; ants 

do not inhabit mounds in 

other regions of the country 

and evidence of previous 

occupations has not been 

found (Melton, 1954). 

 Some authors have 

suggested that mounds result 

from animal disturbance of 

an area. Dalquest and 

Scheffer (1942) and Scheffer 

(1958) asserted that, over 

long periods of time, pocket 

gophers; both recent and 

historic, form pimple hills. 

Ross et al. (1968) extended 

this view to include toads, 

ground squirrels and badgers 

as mound constructors. The 

primary basis for this 

proposed origin was, in each 

instance, the modified soil 

characters, i.e., lower bulk 

density, lack of soil 

structure, increased water 

permeability and particle 

size in mounds, when 

compared to the surrounding 

prairie. These authors 

maintain that these changes 

are possible with the normal 

activities of the suggested 

animals. Grant (1948) 

systematically discredits 

this proposal; his rejection 

of the hypothesis is based 

on the subjective 

interpretation of the pocket 

gopher behavior and the 

inconsistencies in 



44                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

Scheffer’s assertions. It 

seems reasonable that, 

rather than constructing a 

mound to avoid a high water 

table, as Ross et al. (1968) 

suggest, rodents utilized an 

existing mound structure, 

formed by some other means. 

 A series of physical 

explanations for the 

creation of pimple mounds 

has been advanced, including 

suggestions that the mounds: 

are coppices or sand dunes; 

result from fossil mud 

lumps; are protected from 

erosion by cap rock; are the 

work of glaciers; are 

concretionary depositions of 

minerals from ground water; 

results from spring and gas 

vents. Melton (1954) 

discussed these and other at 

length, discrediting each as 

an origin which is generally 

applicable, though each may 

have influenced mound 

formation in a specialized 

locale. 

 Two physical hypotheses 

of pimple mound origin are 

noted as widely acceptable; 

both are based on the 

observation that pimple 

mounds are found in regions 

with at least a moderate 

rainfall. Krinitsky (1949) 

proposed that mounds are 

deposited by river currents 

during periods of high 

water, forming ridges; 

vegetation invades when the 

water recedes, establishes a 

soil and maintains the 

mound. This proposed origin 

is supported by observations 

that mounds lack the well 

defined soil profile common 

in prairies, and that mounds 

are often found along 

ridges; however, Cain (1974) 

asserted that mounds do not 

occur along creeks or rivers 

or in alluvium.  

 Erosion is one of the 

simpler and more popular 

explanations for soil mounds 

(LeConde, 1974); Melton, 

1929); Knechtel, 1952, Cain, 

1974). Knechtel (1952) 

proposed that erosion is 

preceded and enhanced by the 

division of soil into 

prismatic blocks, due to 

seasonal desiccation and 

freezing. The blocks would 

then be worn down and 

rounded by erosion to 

produce pimple mounds. 

Melton (1954) suggested that 

weak, sandy soil erodes 

readily to produce gullies 

with walls which collapse 

easily and are rounded by 

rain and slumping to produce 

mounds. Aronow (1972) 

favored a two-phased 

formative process; firstly, 

a pluvial period marked by 

high run-off to initiate 

mounds; secondly, a period 

of low rainfall and high 

local winds, so that 

vegetation on the mounds 

traps the blowing soil, thus 

creating the thick A-horizon 

which is characteristic of 

pimple hills. Cain (1974) 

postulated that erosion 

around tree cover produces 

pedestal trees; with demise 

of the trees, the eroded 

tree pedestals remain, 

forming small pimple mounds. 



Oklahoma Native Plant Record          45   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

Melton (1954) concluded that 

one is brought by default to 

erosion as the creative 

force behind pimple hills, 

since none of the other 

proposed phenomena is 

sufficiently widespread or 

continuous to account for 

the number of pimple mounds. 

 The vegetation of 

pimpled prairies has been 

examined and analyzed by 

several workers who provide 

varying descriptions. Melton 

(1954) discussed and 

concurred with Campbell’s 

(1906) observations; both 

ascribed to intermounds a 

greater fertility, evidence 

by a darker color, higher 

water content, and higher 

stands of vegetation when 

compared to mounds. 

 McGinnies (1960) took 

the opposing view that 

mounds are more fertile. He 

documented this assertion 

with experimental results of 

the herbage production of 

five species of grasses on 

mounds and intermounds in 

Colorado. McGinnies found 

that mounds yielded twice 

the dry weight herbage of 

the intermounds. Ross et al. 

(1968)) observed that 

prairie mounds were 

dominated by either shrubs, 

forbs, or grasses, depending 

on the size of the mound and 

the amount of soil 

disturbance by animal 

burrowing. Barclay (1938) 

reported some difference in 

the species of plants which 

occur on the mounds and 

intermounds in southeastern 

Oklahoma; notably, the genus 

Drosera occurs only between 

the mounds. 

 The purpose of this 

investigation was to study 

the effect of pimple mound 

microrelief on vegetation, 

particularly on species 

composition and on 

percentage cover. The study 

site selected is a tall 

grass prairie in the 

Cherokee Prairie Biotic 

District of Blair and 

Hubbell (1938). The prairie 

is approximately two miles 

West of Inola and four miles 

East of the Verdigris River 

on Oklahoma Highway 33, NW 

1/4, S6, T 19N, R17E, Rogers 

County, Oklahoma (Fig. 5). 

At the time of the study the 

prairie was owned by K. V. 

Spainhower, who assured the 

author that for more than 

sixty years the prairie had 

been mowed but neither 

grazed nor plowed. This 

prairie was thus deemed to 

be in an undisturbed state, 

when compared to other sites 

available for study, and 

particularly amenable to a 

vegetation study.



46         Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

    

Murray, C.L. 

 

F
i
g
u
r
e
 
5
 
 
L
o
c
a
t
i
o
n
 
o
f
 
s
t
u
d
y
 
s
i
t
e
 
w
i
t
h
 
r
e
f
e
r
e
n
c
e
 
t
o
 
t
h
e
 
t
o
w
n
 
o
f
 
I
n
o
l
a
 
a
n
d
 
t
h
e
 
V
e
r
d
i
g
r
i
s
 
R
i
v
e
r
.
 
U
n
i
t
e
d
 
S
t
a
t
e
s
 

G
e
o
l
o
g
i
c
a
l
 
S
u
r
v
e
y
 
T
o
p
o
g
r
a
p
h
i
c
 
M
a
p
.
 



Oklahoma Native Plant Record          47   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

METHODS AND MATERIALS 
 

 The vegetation of 

mounds and intermounds was 

analyzed by the inclined 

point quadrat method (Tinney 

et al., 1937) using the 

basal contact modification 

(Whitman and Siggeirsson, 

1954). The method was 

applied to each pimple mound 

in the following manner. The 

point frame was placed 

halfway up the eastern slope 

of a mound and readings 

taken from the ten pins; 

then the frame was moved 

five paces in a clockwise 

direction, again placed on 

the mound slope and read. 

This procedure was repeated 

ten times for a total of 

one-hundred points per 

mound. The different 

positionings of the frame 

formed a pattern which 

resembled the spokes on a 

wheel. 

 Similarly, data were 

gathered from the intermound 

region, which concentrically 

surrounded each mound 

considered. The frame was 

again positioned ten times, 

at ten pace intervals, 

providing one-hundred points 

per intermound region 

sampled.  

 Data were collected 

from eighteen mounds and 

intermound areas using the 

point frame, in April, June 

and July, 1972. The mounds 

considered were selected 

randomly. These data were 

tabulated and used to 

determine percentage cover 

by vegetation. 

 Vascular plants of the 

study site were collected 

weekly, from mid-march to 

mid-July, 1972; when 

possible, plants were 

collected while blooming. 

Nomenclature follows 

Waterfall (1969). Upon 

collection, it was noted 

whether each species 

occurred on mounds, 

intermounds, or both. All 

plant specimens are 

contained in The University 

of Tulsa Herbarium.  

 The spatial 

relationship between the 

mounds on the prairie was 

mapped. The fence marking 

the southern boundary of the 

study site was designated as 

the primary reference line; 

mounds were mapped in 

relationship to this fence. 

The information for the map 

was obtained using a Brunton 

compass, corrected to true 

North. The heights of the 

mounds were measured in 

relation to each other and 

to the intermound surface. 

The slope of the intermound 

surface was also measured 

with the compass, and 

distances between the mounds 

with a metal tape.  

 Soil samples were 

collected from the surface 

to bedrock, from three 

mounds and corresponding 

intermound areas, using a 

hydraulic soil auger two 

inches in diameter. The 

auger was provided and 



48                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

operated by Doc Polone of 

the Rogers County Soil 

Conservation  

Service, Claremore, 

Oklahoma. Soil cores 

collected in this manner 

were placed intact in 

troughs made of plastic 

pipe; the pipes, 2.5 inches 

(6.4 cm) in diameter, had 

been split lengthwise to 

accommodate the samples. 

Soil horizons were 

identified as definitively 

as possible, considering the 

color and macroscopic 

character of the soil. Fifty 

grams of soil from each 

horizon was characterized as 

to its physical composition, 

percentages of sand, silt 

and clay, by the hydrometer 

method (Bouyoucos, 1936). 

 The climatological data 

presented were provided by 

the United States Weather 

Bureau Office at Tulsa 

International Airport. 

Precipitation is based on 

records dating back to 1950 

(Curry, 1970). 

 The statistical tests 

used were the point biserial 

correlation to analyze the 

percent cover data (Downie 

and Heath, 1974); and the 

two-factor mixed design 

analysis of variance to 

analyze the data relating to 

the physical composition of 

the soils (Winer, 1962).  

 

RESULTS 
 

 Seventy-six species of 

vascular plants were 

collected from the study 

site. This collection 

represents a total of 

twenty-eight families and 

sixty-three genera; species 

are listed phylogenetically 

in Table I. From this 

collection, twenty-five 

species (33%) were found 

only in mound habitats, 

twenty-nine species (38%) 

were collected only from 

intermound habitats, and 

twenty-two species (29%) 

occurred in both habitats. 

This relationship between 

species and habitat is 

presented in Table II. Of 

the species common to both 

habitats in the prairie, 

seven species (33%) were 

grasses; one-half the 

grasses were restricted to 

one habitat or the other.  

 Percent cover of the 

prairie by living vegetation 

in the two habitats is 

compared in Fig. 6. The 

percent cover value 

presented for each date is 

the mean value for the data 

collected on that date. By 

inspection, the data reveal 

that higher values of 

percent cover were obtained 

for mounds than for 

intermounds. This 

information was submitted to 

the point biserial 

statistic. The t statistic 

indicated that the 

relationship between the 

type of terrain, i.e., mound 

or intermound, and the 

percent cover values was not 

statistically significant t 

= 1.64, df = 32, p < .10). 

 The low precipitation 

for the period preceding and 



Oklahoma Native Plant Record          49   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

during this study in 1972 is 

presented and compared with 

that of the previous twenty 

years in Fig. 7. 

 The spatial 

relationship between the 

pimple mounds on the prairie 

considered in this study is 

shown in Fig. 8. The mounds 

were found primarily along a 

ridge, from which the 

prairie sloped westwardly 

with approximately 10% 

slope. In this prairie the 

mounds ranged from 16 feet 

(4.876 m) to 50 feet (15.24 

m) in diameter and from 1.5 

feet (0.457 m) to 3.0 feet 

(0.914 m) in height. Mounds 

varied in shape from 

irregularly eggshaped [40 

feet(12.19 m) by 50 feet 

15.24 m)] to nearly circular 

[26 feet (7.924 m) by 27 

feet (8.229 m)]. There was 

generally noted a slight 

alignment down slope. 

 Soil depth and the 

approximate thickness of 

each horizon from three 

mounds and corresponding 

intermound areas are 

compared in Fig. 9. The 

physical composition of the 

soil from each horizon is 

summarized in Table III. The 

difference in the amount of 

sand in the three horizons 

was statistically 

significant (F = 6.999; df = 

2.8; p <.02). Horizon C had 

less sand than either A or 

B. The difference in the 

amount of silt at the three 

horizons was statistically 

significant (F = 47.794; df 

= 2.8; p < .001). Horizon A 

had more silt than either B 

or C. The difference in the 

amount of clay in the three 

horizons was statistically 

significant (F = 7.335; df = 

2.8; p < .02). Horizon C had 

more clay than either A or 

B. When comparing soil from 

mounds with that of 

intermound areas, the 

difference in the 

percentages of sand, silt 

and clay was not 

statistically significant 

(Sand: F = 1.813; df = 1.4; 

p < .25), (Silt: F = 3.460; 

df = 1.4; p < .15), (Clay: F 

= 1.041; df = 1.4; p < .4). 

 

 

 

 

 

 
 

Figure 6  The percentage of cover by 

living vegetation of mounds (O), and 

intermounds (<>), determined with a 

point frame, basal contact 

modification. 

  

 



50                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

 
 

Figure 7   Monthly precipitation 

figures for 1972 (x), compared with 

monthly means calculated from 1951-

1970 (o) for Tulsa County. From U.S. 

Weather Bureau. 

 

 

 

 

 

 
 
Figure 9   Comparison of soil depths 

and soil horizons for three mounds and 

corresponding intermound regions. 

 

 

 
 

Figure 8  Distribution of mounds in 

the prairie studied. Note that the 

distribution pattern generally follows 

the higher contour lines. (Scale 

1:17,352)  

 

 

DISCUSSION 
 

 One factor to be 

considered to interpreting 

the data presented is the 

distribution of the rainfall 

for the duration of the 

experiment; for this period, 

the precipitation was lower 

than average which produced 

a marked effect on the 

vegetation. Many plants 

appeared vegetatively but 

growing tip shriveled and 

turned brown before flowers 

bloomed; therefore, many of 

these plants were not 

identified. It is possible 

that the low rainfall for 

this early summer period 

reduced the number of 

species recorded for the 

prairie. 



Oklahoma Native Plant Record          51   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

 In both mound and 

intermound habitats the 

small forbs dried in May and 

June, leaving a vegetation 

consisting mainly of 

grasses. Although this 

phenomenon is common in 

prairies, it usually occurs 

later in the growing season 

and after plant reproductive 

cycles have been completed. 

This early desiccation of 

the prairie and its 

vegetation is reflected in 

the percent cover figures, 

which steadily decreased 

through June until cover was 

recorded as one percent in 

mid-June (Fig. 6). 

Generally, the intermound 

vegetation evidenced 

desiccation before that on 

the mounds. The low rainfall 

may have accentuated the 

difference recorded in mound 

and intermound vegetation 

cover, which will be 

discussed later. The 

rainfall is probably 

partially responsible for 

the low range of percent 

cover when these figures are 

compared to those of other 

studies (Drew, 1944); 

Whitman and Siggeirsson, 

1954). 

 The method of plant 

collection rendered the 

information unsuitable for 

statistical analysis of the 

interaction between species 

occurrence and habitat 

(Table II). Upon inspection, 

there is no obvious 

distribution of certain 

families to either mounds or 

intermound regions. It is 

noted that some genera 

within a family and some 

members of a genus were 

distributed, one in mound 

habitats, another in 

intermound habitats. The 

distribution, as recorded, 

reflects the interaction of 

the microhabitat conditions 

most favorable to each 

species. Determination of 

these conditions would 

require extensive research 

on the physiological growth 

requirements for each 

species. As mentioned above, 

some species never bloomed, 

due to drought; perhaps in a 

more pluvial year, when a 

greater number of 

representatives from each 

family bloomed, more 

conclusive remarks could be 

made about the distribution 

of plant families in the 

mound and intermound 

habitats. 

 Analysis of the 

difference in the percent 

cover data of mounds and 

intermounds revealed this 

difference to be 

statistically insignificant 

(Fig.6). Thus it is 

concluded that the variation 

of the percent cover was 

possible due to chance 

alone, and that the 

vegetation on mounds and 

intermounds is one plant 

population, not two. The 

inclined point quadrat 

method was used in this 

study with the basal contact 

modification, attempting to 

reduce the quantity of data 

and yet provide an accurate 



52                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

assessment of the percent 

cove in the two habitats. 

Perhaps it would be possible 

to arrive at the conclusion 

that the two habitats 

support different 

populations, if the inclined 

point quadrat method were 

used without the basal 

modification. This would 

provide indirect information 

on the herbage production of 

the two habitats, in 

addition to the information 

collected with the basal 

contact modification. It was 

observed that the vegetation 

on the mounds was more 

luxuriant than in intermound 

areas (Fig. 2); perhaps this 

modification of the 

experiment as performed 

would accentuate the 

difference recorded into one 

significant both 

statistically and 

biologically. 

 Summarizing the soil 

analysis data, mounds and 

intermounds did not vary 

significantly in physical 

composition. Alternatively, 

it is possible that the 

soils varied chemically, but 

such determinations were not 

made. The soils of mounds 

and intermounds did vary in 

depth (Figs. 9 and 10). The 

difference noted in the 

percent cover on mounds and 

intermounds may possibly be 

attributed to the available 

water held in each soil. 

Perhaps the deeper mound 

soil provided for more 

extensive root growth, a 

cooler soil, and greater 

soil area for storage of 

available water; in 

response, mound vegetation 

exhibited a somewhat higher 

percent cover and particular 

species composition. 

Conversely, the soil depth 

of intermounds allowed less 

root growth; provided a 

warmer soil and less soil 

space for water storage; 

thus intermound vegetation 

was lower in percent cover 

and contained species 

adapted to these conditions. 

If soil depth and its 

corresponding available 

water are influential in 

determining vegetation, it 

is possible that the 

difference noted in mound 

and intermound vegetation 

was greater this dry year 

than might normally be the 

case. 

 One difference between 

mounds and intermounds which 

may be significant is that 

most mounds were observed to 

contain the burrows of 

numerous animals; skunks, 

field mice and snakes were 

observed to inhabit mounds. 

The presence of these 

animals could influence the 

vegetation in numerous ways, 

by providing organic 

fertilizer to the mounds, 

changing the carbon dioxide 

levels of the mound soil, 

varying the compaction of 

the soil and displacing the 

root systems, especially tap 

roots. Depending on the 

extent of animal 

disturbance, the mounds 

could provide a habitat 



Oklahoma Native Plant Record          53   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

which varied greatly from 

intermound habitats, a 

difference which could 

easily be mirrored by 

vegetation. 

 The observations and 

data support some of the 

general ideas offered in the 

introduction concerning 

pimple hills and refute 

others. The mound soil 

considered here was 

definitely not loess-like, 

but a sandy loam. The soil 

of mounds and intermounds 

was physically similar, and 

revealed similar horizons. 

There was no evidence to 

suggest that mounds were 

formed by animal activity, 

though animals were present 

on the study site. That the 

mounds studied occurred on 

upland ridges, were composed 

of sandy soil, and were 

aligned downslope, support 

the hypothesis that erosion 

creates and maintains 

mounds. Whether this erosion 

was enhanced by frost or 

desiccation fissures or tree 

pedestals were not 

determined. Trees do not now 

occur along the ridges 

considered; whether they 

once initiated mound 

formation might be 

elucidated by sectioning the 

mound to determine residual 

evidence of roots. Such 

sections, which would also 

have aided in estimating the 

extent of animal activity, 

were not made on the 

insistence of the property 

owner. 

 Thus it may be said 

that the prairie considered 

exhibited a microrelief 

feature of pimple mounds and 

intermound regions, but the 

two soils did not vary in 

physical composition. These 

mounds were found to lie in 

an irregular pattern along 

an upland ridge with a 

slight alignment downslope. 

Mounds supported vegetation 

which differed in species 

composition from the 

vegetation of intermound 

regions; there was no 

significant difference in 

the percent cover by 

vegetation of these two 

prairie habitats, though the 

unassessed data suggested 

that the mounds were more 

productive than the 

intermound prairie. The 

abnormally low rainfall 

through the experimental 

period doubtlessly 

influenced the results of 

the vegetation study. 

Perhaps repetition of the 

vegetation study, with the 

modifications proposed 

herein, during a growing 

season with more normally 

distributed precipitation, 

would produce more reliable 

results, providing for more 

conclusive remarks about the 

vegetation of mounds and 

intermounds than can be 

extended here. 

 



54                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

 

TABLE 1 

LIST OF THE VASCULAR PLANTS COLLECTED FROM MARCH THROUGH 

JULY, 1972, AT THE STUDY SITE LISTED BY FAMILIES 

(Nomenclature according to U.T. Waterfall (1969); 

common names according to Gleason (1963) 

and Rechethin (1954) 

 

 Scientific Name      Common Name 

 

Graminae 

 Andropogon Gerardi    Big blue stem 

 Andropogon scoparius    Little blue stem 

 Andropogon ternaries    Splitbeard blue stem 

 Bouteloua curtipendula   Side-oats grama 

 Bromus japonicus     Japanese brome grass 

 Cynodon Dactylon     Bermuda grass 

 Festuca octoflora    Sixweeks fescue 

 Hordeum pusillum     Little barley 

 Lolium multiflorum    Italian rye grass 

 Manisuris cylindrica    Carolina jointtail 

 Panicum oligosanthes var.      

  Scribnerianum    Scribners panicum 

 Panicum Ravenelii    -- 

 Panicum sphaerocarpon    Roundseed panicum 

 Panicum virgatum     Switchgrass 

 Sorghastrum nutans    Indian grass 

 Sporobolus cryptandrus   Sand dropseed 

Cyperaceae 

 Carex caroliniana    Carolina sedge 

 Cyperus filiculmis    Slenderleaf sedge 

Liliaceae 

 Camassia scilloides    Wild hyacinth, Atlantic 

         camas 

 Erythronium albidum    Dogtooth violet,  

         white fawn lily 

 Nothoscordum bivalve    Yellow false-garlic 

Amaryllidaceae 

 Hypoxis hirsute     Stargrass, common  

         goldstar 

Orchidaceae 

 Spiranthes vernalis    Upland ladies tresses 

Santalaceae 

 Commandra Richardsoniana   Bastard toad-flax 

Polygonaceae 

 Eriogonum longifolium    Longleaf wild-buckwheat 

Portulacaceae 

 Claytonia virginica    Spring beauty 

Caryophyllaceae 

 Arenaria patula     Pitchers sand wort 

 Cerastium vulgatum var.   Mouse-ear chickweed, 

  vulgatum     Big chickweed 



Oklahoma Native Plant Record          55   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

Ranunculaceae 

 Anemone virginiana    Virginia anemone, wind-

         flower 

 Delphinium tricornis forma     

  albiflora     Rock larkspur 

 Ranunculus hispidus    Bristly buttercup 

Cruciferae 

 Selenia aurea     Yellow selenia 

Saxifragaceae 

 Saxifraga texana     Texas saxifrage 

Leguminoseae 

 Dalea purpurea     -- 

 Lotus americanus     Deervetch 

 Schrankia uncinata    Catclaw sensitive briar 

 Tephrosia virginiana var.    Goats rue, Virginia  

  holosericea     tephrosia 

Oxalidaceae 

 Oxalis corniculata    Wood sorrel, creeping 

         oxalis 

 Oxalis dillenii     Sheep sorrel, wood  

         sorrel 

Polygalaceae 

 Polygala incarnate    Pink milkwort 

 Polygala sanguinea    Blood milkwort 

Umbelliferae 

 Eryngium yuccafolium var. 

  synchaetum     Yucca-leafed eryngo 

 Polytaenia Nuttallii var.      

  Nuttallii     Prairie parsley 

Gentianaceae 

 Sabatia campestris forma      

  campestris     Prairie rosegentian 

 

Asclepiadaceae 

 Asclepias stenophylla    Slimleaf milkweed 

 Asclepias veridis    Milkweed 

Hydrophyllaceae 

 Phacelia strictiflora    Prairie phacelia 

Labiatae 

 Scutellaria parvula var.  

  Leonardi     small skullcap  

Solanaceae 

 Solanum carolinense forma    

  carolinense    Carolina horse nettle 

Scrophulariaceae 

 Buchnera Americana    American blueheart 

 Castelleja coccinea var.   Painted cup, Indian 

   coccinea     paintbrush 

 Linaria canadensis var.      

  texana     Old-field toadflax 

 Penstemon tubaeflorus    Beard tongue, tube  

         penstemon 



56                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

Acanthaceae 

 Ruellia humilis     Low ruellia 

Plantaginaceae 

 Plantago aristata    Bottlebrush plantain 

 Plantago elongate    -- 

 Plantago media     -- 

 Plantago Purshii var. Purshii  Wooly plantain 

 Plantago virginica    Paleseed plantain 

Rubiaceae 

 Hedyotis crassifolia    Tiny bluet 

Valerianaceae 

 Valerianella Nuttallii   Nuttall cornsalad 

Campanulaceae 

 Lobelia spicata var. 

  leptostachys    Palespike lobelia 

 Specularia perfoliata    Clasping Venus’   

         lookingglass 

Compositae 

 Achillea lanulosa    Western yarrow 

 Antennaria neglecta   Everlasting, pussytoes 

 Aster ericoides    Wild aster, heath aster 

 Coreopsis grandiflora   Tickseed, big-flower   

        coreopsis 

 Echinacea pallida   Coneflower, pale echinaceae 

 Erigeron strigosus   Daisy fleabane,  

        prairie fleabane 

 Erigeron tenuis    Slender fleabane 

 Gnaphalium purpureum   Cudweed, everlasting,  

        purple cudweed 

 Hieracium longipilum   Longbeard hawkweed 

 Krigia dandelion    Tuber dwarf dandelion 

 Liatris pychnostachya   Blazing star, Kansas   

        gayfeather 

 Rudbeckia hirta    Coneflower, blackeyed susan 

 Solidago mollis    Ashy goldenrod 

 



Oklahoma Native Plant Record          57   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

TABLE II 

PLANTS COLLECTED, NOTING WHETHER THE SPECIES OCCURRED  

ON MOUNDS, IN INTERMOUND REGIONS, OR BOTH 

(ASTERISK DESIGNATES IN WHICH HABITAT A SPECIES FIRST BLOOMED) 
 

Mound Species 
 

Achillea lanulosa forma lanulosa   Panicum Ravenelii 

Bromus japonicus      Panicum virgatum 

Cerastium vulgatum var. vulgatum   Phacelia strictiflora 

Claytonia virginica      Plantago aristata 

Delphinium tricorne forma    Plantago elongate 

 albivlora      Plantago Purshii var. 

Echinacea pallida       Purshii 

Erigeron strigosus     Rudbeckia hirta 

Eriogonum longifolium     Solanum carolinense forma 

Hieracium longipilum      carolinense 

Lolium multiflorum     Solidago mollis 

Manisuris cylindrical     Specularia perfoliata 

Oxalis dillenii      Tephrosia virginiana var. 

Panicum oligosanthes var.     holosericea 

 Scribnerianum     Valerianella Nuttallii 
 

Intermound Species 
 

Antennaria neglecta     Lotus americanus 

Arenaria patula      Nothoscortum bivalve 

Asclepias stenophylla     Oxalis corniculata 

Bouteloua curtipendula     Panicum sphaerocarpon 

Bucchnera Americana     Penstemon tubaeflorus 

Camassia scilloides     Plantago media 

Castilleja coccinea forma coccinia   Plantago virginica 

Dalea purpura      Polygala incarnate 

Erigeron tenuis      Polygala sanguinea 

Eryngium yuccafolium var.    Polytaenia Nuttallii var. 

 synchaetum       Nuttallii 

Erythronium albidum var. albidum   Saxifraga texana 

Hordeum pusillum      Scutellaria parvula var. 

Hypoxis hirsute       Leonardi 

Krigia dandelion      Selenia aurea 

Liatris pychnostachya     Spiranthes vernalis 
 

Mound and Intermound Species 
 

Andropogon Gerardi     Gnaphalium purpureum 

Andropogon scoparius     Hedyotis crassifolia 

Andropogon ternaries     Linaria canadensis var. 

Anemone virginiana *M      texana *IM 

Asclepias viridis      Lobelia spicata var. 

Aster ericoides *IM      leptostachys 

Carex caroliniana      Ranunculus hispidus 

Comandra Richardsoniana     Ruellia humilis 

Coreopsis grandiflora     Sabatia campestris forma 

Cynodon dactylon       campestris 

Cyperus filiculmis     Schrankia uncinata 

Festuca octoflora *IM     Sorghastrum nutans 

        Sporobollus cryptandrus 



58                             Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

            
   

Murray, C.L. 

 

TABLE III 

COMPARISON OF THE PHYSICAL COMPOSITION OF THE SOIL 

FROM MOUND AND INTERMOUND REGIONS, DETERMINED BY THE BOUYOUCOS METHOD 

(MOUNDS AND INTERMOUNDS ARE DESIGNATED BY Numbers 1,2,3.  

Soil horizons are noted as A, B, C) 

 

MOUND    Average     Average   Average 

 % Sand % Sand % Silt % Silt % Clay % Clay 

1A 85.2  10.7  4.1 

 

 

2A 78.4 80.5 14.8 13.7 6.8 5.8 

 

3A 77.9  15.6  6.5 

 

 

1B 85.7  9.4  4.9 

 

 

2B 79.4 83.4 9.7 8.9 10.9 7.7 

 

3B 85.2  7.6  7.2 

 

 

1C 79.0  11.3  9.7 

 

 

2C 71.4 71.5 11.6 11.4 17.0 17.1 

 

3C 64.2  11.3  24.5 

 

 

 

 

INTERMOUND   Average   Average  Average 

 % Sand % Sand % Silt % Silt % Clay % Clay 

1A 79.6  11.5  8.9 

 

 

2A 78.4 77.9 15.3 14.9 6.3 

 

7.2 

3A 75.7  17.9  6.4 

 

 

1B 82.0  12.5  5.5 

 

 

2B 71.6 75.3 15.4 15.9 13.0 8.8 

 

3B 72.2  19.7  8.1 

 

 

1C 81.8  11.3  6.9 

 

 

2C 50.8 58.5 13.9 14.3 35.5 

 

27.2 

3C 42.8  17.8  39.4 

 

 

 

 



Oklahoma Native Plant Record          59   
Volume 5, Number 1, December 2005  

  
 

  Murray, C.L. 

LITERATURE CITED 
 

Aronow, S. 1972. Personal 

communication. 

Barclay, H.G. 1938. A 

preliminary report of 

the ecology of a 

Drosera meadow. Proc. 

Okla. Acad. Sci. 

18:22-25. 

Blair, W.F. and T.H. 

Hubbell. 1938. The 

biotic districts of 

Oklahoma. Am. Midland 

Naturalist. 20:425-

454. 

Bouyoucos, G.J. 1936. 

Directions for making 

mechanical analyses of 

soils by the 

hydrometer method. 

Soil Sci. 42:225-229. 

Branner, J.C. 1905. Natural 

mounds or “hog 

wallows”. Science, n. 

ser. 21:514-516. 

Cain, R.H. 1974. Pimple 

mounds: a new 

viewpoint. Ecology 

55(1):178-182. 

Campbell, M.R. 1906. Natural 

mounds. J. Geol. 

14:707-714 

Curry, B.R. 1970. Climate of 

Oklahoma. U.S. Dept of 

Commerce Climatography 

of the United States 

Nol. 60-34. Rev. ed. 

U.S. Govt. Printing 

Office, Washington, 

D.C. 

Dalquest, W.W., and V.B. 

Scheffer. 1942. Origin 

of the Mima mounds of 

western Washington. J. 

Geol. 50:68-84. 

Drew, W.B. 1944. Studies on 

the use of the point 

quadrat method of 

botanical analysis of 

mixed pasture 

vegetation. J. Agr. 

Res. 69:289-297. 

Downie, N.M., and R.W. 

Heath. 1974. Basic 

statisticsl methods. 

4
th
 ed. Harper and Row, 

Publishers, New York, 

Evanston, San 

Francisco and London. 

335 p. 

Fenneman, N.M. 1931. 

Physiography of 

eastern United States. 

McGraw-Hill Book 

Company, New York. 714 

p. 

Gleason, H.A. 1963. The new 

Britton and Brown 

illustrated flora of 

the northeastern 

United States and 

adjacent Canada. 3
rd
 

print., slightly rev. 

Hafner Publishing 

Company, Inc., New 

York and London. 3 v. 

Grant, C. 1948. Mima mounds. 

J. Geol. 56:229-231. 

Knechtel, M.M. 1952. Pimpled 

plains of eastern 

Oklahoma. Geol. Soc. 

Amer. Bull. 63:689-

700. 

Krinitsky, E.L. 1949. Origin 

of pimple mounds. Am. 

J. Sci. 247: 706-714. 

LeConde, J. 1874. On the 

great lava flood of 

the Northwest and on 

the structure and age 

of the Cascade Mts. 



60 Oklahoma Native Plant Record 
Volume 5, Number 1, December 2005 

Murray, C.L. 

Am. J. Sci. Ser. 

3,7:259-367. 

McGinnies, W. 1960. Effect 

of Mima-type 

microrelief on herbage 

production of five 

seeded grasses in 

western Colorado. J. 

Range Management. 

13(5):213-234. 

Melton, F.A. 1929. Natural 

mounds of northeastern 

Texas, southern 

Arkansas, and northern 

Louisiana. Proc. Okla. 

Acad. Sci. 9:119-130. 

___________. 1954. Natural 

mounds of northeastern 

Texas southern 

Arkansas and northern 

Louisiana. The Hopper, 

Oklahoma Geol. Surv. 

14:88-121. Privately 

published. 

Prescott, J.A. 1931. The 

soils of Australia in 

relation to vegetation 

and climate. 

Commonwealth of 

Australia, Council for 

Scientific and 

Industrial Research 

Bull. 52. 71 p. 

Rechethin, C.A. 1954. A 

guide to plant names 

in Texas, Oklahoma, 

Louisiana and 

Arkansas. Rev. ed. 

U.S. Dept. Agr., Soil 

Conservation Service, 

Fort Worth. 91 p. 

Ross, B.A., J.R. Tester, and 

W.J. Breckenridge. 

1968. Ecology of Mimi-

type mounds in 

northwestern 

Minnesota. Ecology 

49(1):172-177. 

Scheffer, V.B. 1958. Do 

fossorial rodents 

originate Mima-type 

microrelief. Am. 

Midland Naturalist. 

59:505-510. 

Tinney, F.W., O.S. Aamodt, 

and H.L. Ahlgren. 

1937. Preliminary 

report of a study on 

methods used in 

botanical analysis of 

pasture swards. J. Am. 

Soc. Agr. 29:835-840. 

Veatch, A.C. 1906. On 

thehuman origin of 

small mounds of the 

lower Mississippi 

Valley and Texas. 

Science. 23(575):34-

36. 

Waterfall, U.T. 1969. Keys 

to the flora of 

Oklahoma 4
th
 ed.

Stillwater. Privately 

published.  

Whitman, W.C., and E.I. 

Siggeirsson. 1954. 

Comparison of line 

intercept and point 

contact methods in the 

analysis of mixed 

grass range 

vegetation. Ecology 

35:431-436. 

Winer, B.J. 1962. 

Statistical principles 

in experimental 

design. McGraw-Hill 

Book Company, New 

York. Chapter 7. p 

302-313.