Pearce-Sellards 
Series 


45 


EVOLUTION 
OF 
THE 
GENUS 
HOLMESINA 
(PAMPATHERIIDAE, 
MAMMALIA) 


IN 
FLORIDA, 
WITH 
REMARKS 
ON 
TAXONOMY 
AND 
DISTRIBUTION. 


A. 
Gordon 
Edmund 
MAY 
28,1987 
TEXAS 
MEMORIAL 
MUSEUM, 
THE 
UNIVERSITY 
OF 
TEXAS 
AT 
AUSTIN 



A. 
Gordon 
Edmund 
is 
Curator 
of 
Vertebrate 
Paleontology 
at 
the 
RoyalOntario 
Museum, 
Toronto, 
and 
Associate 
Professor 
of 
Geology 
at 
the 
University 
ofToronto. 
Editor 
-Georg 
Zappler 


-

Layout. 
Typesetting 
Mariah 
Mayfield 


The 
Pearce-Sellards 
Series 
is 
an 
occasional, 
miscellaneous 
series 
of 
brief 
reports 
of 
Museum 
and 
Museum-associated 
field 
investigations 
and 
other 
research. 
All 
manuscripts 
are 
subjected 
to 
extramural 
peer 
review 
before 
being 
accepted. 
The 
series 
title 
commemorates 
the 
first 
two 
directors 
of 
Texas 
Memorial 
Museum, 
both 
now 
deceased: 
Dr. 
J. 
E. 
Pearce, 
Professor 
of 
Anthropology, 
and 
Dr. 
E. 
H. 
Sellards, 
Professor 
of 
Geology, 
The 
Universityof 
Texas 
at 
Austin. 


© 
1987 
by 
Texas 
Memorial 
Museum 
The 
University 
of 
Texas 
at 
Austin 
All 
rights 
reserved 
Printed 
in 
the 
United 
States 
of 
America 



CONTENTS 


Abstract, 
in 
English 
and 
Spanish 
1 
Acknowledgements 
2 
Abbreviations 
and 
Acronyms 
2 
Key 
Words 
2 
Introduction 
3 
Material 
3 
History 
of 
North 
American 
Collecting 
4 
Taxonomic 
History 
Pertinent 
to 
North 
America 
5 
Summary 
of 
Pampathere 
Systematics 
7 
Increasing 
Size 
ofHolmesina 
in 
Florida 
9 
Carapace 
9 
Limb 
Bones 
11 
Metapodials 
15 
Taxonomic 
Conclusions 
15 
References 
18 



LIST 
OF 
FIGURES 


Fig. 
1. 
Two 
buckler 
and 
one 
imbricating 
osteoderms 
of 
each 
of 
three 
species 
of 
pampatheres. 
A) 
ROM:B 
7960 
Holmesina 
floridanus, 
Blancan, 
El 
Jobean, 
Charlotte 
Co., 
Florida. 
B) 
FSM 
9336 
Holmesina 
septentrionalis, 
Rancholabrean, 
Branford, 
Suwanee 
Co., 
Florida. 
C) 
ROM 
23829 
Kraglievichia 
paranense, 
Montehermosan, 
Entre 
Rios, 
ArgentinaFig. 
2. 
Left 
calcanea 
of 
Floridean 
species 
ofHolmesina. 
A) 
RHB 
3151 
H. 
septentrionalis, 
Rancholabrean,
Sarasota 
Co. 
B) 
FSM 
20954 
Holmesina 
sp., 
Irvingtonian,
Haile 
XVI 
A 
(reversed 
from 
right). 
C) 
FSM 
20946 
Holmesina 
sp., 
Irvingtonian, 
Inglis 
I 
A. 
D) 
ROM 
21803 
H. 
floridanus. 
Blancan, 
Haile 
XV 
A 
(cast)
Fig. 
3. 
Relation 
between 
metacarpal 
length 
and 
time 
in 
Holmesina 
from 
Florida 
Fig. 
4. 
Relation 
between 
metatarsal 
length 
and 
time 
in 
Holmesina 
from 
Florida 
10 
14 
16 
17 
LIST 
OF 
TABLES 
Table 
1. 
Increase 
in 
osteoderm 
size 
between 
Blancan 
and 
Rancholabrean 
samplesTable 
2. 
Increase 
in 
bone 
lengths 
from 
Blancan 
to 
Rancholabrean 
times 
Table 
3. 
Changes 
in 
tibia-fibula 
measurements 
from 
Blancan 
to 
Rancholabrean 
times 
Table 
4. 
Changes 
in 
radius 
dimensions 
from 
Blancan 
to 
Rancholabrean 
times 
Table 
5. 
Changes 
in 
calcaneum 
dimensions 
from 
Blancan 
to 
Rancholabrean 
times 
Table 
6. 
Changes 
in 
the 
ratio 
of 
tuber 
to 
neck 
in 
the 
calcaneum 
from 
Blancan 
to 
Rancholabrean 
times 
Table 
7. 
Changes 
in 
the 
separation 
of 
medial 
and 
lateral 
facets 
on 
the 
calcaneum 
and 
astragalus 
from 
Blancan 
to 
Rancholabrean 
times 
Table 
8. 
Changes 
in 
dimensions 
of 
facets 
on 
the 
calcaneum 
from 
Blancan 
to 
Rancholabrean 
times 
10 
11 
12 
12 
12 
13 
13 
13 



ABSTRACT 


A 
large 
series 
of 
specimens 
of 
pampatheres 
(fossil 
giant 
armadillos) 
has 
been 
found 
in 
Florida, 
covering 
approximately 
two 
million 
years 
from 
their 
emigration 
from 
South 
America 
to 
their 
late 
Pleistocene 
extinction. 
During 
this 
period, 
there 
was 
little 
morphological 
change, 
although 
the 
limb 
bones 
elongated 
by 
about 
50%. 
The 
area 
ofindividual 
osteoderms, 
which 
is 
indicative 
of 
the 
surface 
area 
of 
the 
animal, 
increased, 
as 
expected, 
by 
the 
square 
of 
that 
amount, 
i.e. 
2.2 
times. 


The 
generic 
name 
Kraglievichia 
Castellanos 
1927, 
is 
incorrect 
for 
the 
Floridean 
Blancan 
species, 
as 
is 
Pampatherium 
Ameghino 
1875 
for 
the 
Rancholabrean 
species. 
Holmesina 
Simpson 
1930 
is 
the 
correct 
genus 
for 
the 
entire 
chronocline 
in 
Florida, 
with 
H. 
floridanus 
(Robertson 
1976) 
for 
the 
Blancan 
species 
and 
//. 
septentrionalis 
(Leidy 
1889) 
for 
the 
Rancholabrean 
and 
late 
Irvingtonian 
species. 
Specimens 
of 
Lujanian 
age 
from 
Ecuador, 
Peru, 
and 
Venezuela 
are 
provisionally 
referred 
to 
Holmesina 
occidentalis 
(Hoffstetter 
1948), 
and 
those 
of 
the 
same 
age 
from 
Brasil 
are 
provisionally 
referred 
to 
Holmesina 
humboldti 
(Lund 
1839). 
No 
ancestral 
forms 
are 
currently 
recognized 
among 
either 
the 
published 
or 
unpublished 
material 


from 
South 
America. 


RESUMEN 


Un 
buen 
numero 
de 
especimens 
de 
pampaterios 
(armadillos 
gigantes 
fosiles)
han 
sido 
encontrados 
en 
Florida, 
abarcando 
un 
lapse 
de 
aproximadamentedos 
millones 
de 
ahos, 
desde 
su 
inmigracion 
de 
Sud 
America 
hasta 
su 
extincion 
en 
el 
Pleistocene 
tardi'o. 
Durante 
este 
periodico 
pocos 
cambios 
morfologicos 
se 
presentaron, 
ha 
pesar 
de 
que 
los 
huesos 
apendiuclares 
se 
alagaron 
en 
un 
50%. 
El 
area 
de 
las 
placas 
oseas 
individuals, 
la 
cual 
es 
indicativa 
del 
area 
superficial 
del 
animal, 
aumento, 
como 
era 
esperado, 
al 
cuadrado 
de 
esa 
cifra 
i.e. 
2.2 
veces. 


El 
nombre 
generico 
de 
Kraglievichia 
Castellanos 
1927, 
originalmenteasignado 
a 
la 
especie 
Blanqueana 
de 
Florida 
es 
equivocado, 
como 
lo 
es 
tambien 
Pampatherium 
Ameghino 
1875 
para 
la 
especie 
Rancholabreana. 


Yo 
propongo 
que 
el 
genero 
Holmesina 
Simpson 
1930, 
sea 
reconocido 
como 
valido 
para 
la 
cronoclina 
completa 
de 
Florida; 
con 
H. 
floridanus 
(Robertson 
1976) 
para 
la 
especie 
Blanqueana 
y 
H. 
septentrionalis 
(Leidy1889)para 
la 
Rancholabreana. 


Las 
especies 
del 
Pleistocene 
tardfo, 
H. 
occidental 
(Hoffstetter 
1948)
del 
Ecuador, 
Peru 
y 
Venezuela, 
y 
H. 
humboldti 
(Lund 
1839) 
del 
Brasil 
posiblemente 
puedan 
ser 
referidas 
tambien 
como 
Holmesina. 


Se 
han 
buscado 
posibles 
ancestros 
entre 
el 
material 
sudamericano 
publicado 
y 
no 
publicado, 
pero 
ninguno 
parece 
ser 
eligible. 



ACKNOWLEDGEMENTS 


Special 
thanks 
are 
due 
to 
Dr. 
S. 
David 
Webb 
of 
Florida 
State 
Museum, 
University 
of 
Florida, 
for 
his 
patience 
in 
permitting 
me 
to 
examine 
and 
borrow 
large 
numbers 
of 
specimens 
in 
his 
care. 
1 
also 
wish 
to 
thank 
Dr. 
Clayton 
Rayof 
U.S. 
National 
Museum, 
Dr. 
William 
Turnbull 
of 
the 
Field 
Museum 
of 
Natural 
History, 
Dr. 
Don 
Savage 
of 
the 
Museum 
of 
Paleontology, 
Universityof 
California 
at 
Berkeley, 
Dr. 
Ernest 
Lundelius 
ofTexas 
Memorial 
Museum, 
and 
several 
other 
professional 
paleontologists. 
Special 
mention 
must 
also 
be 
made 
of 
many 
generous 
and 
enthusiastic 
Floridean 
amateur 
paleontologists,
including 
Lelia 
and 
William 
Brayfield, 
Helen 
and 
Roy 
Burgess, 
and 
Ben 
Waller, 
to 
name 
only 
a 
few. 


The 
quality 
of 
the 
manuscript 
was 
greatly 
improved 
by 
the 
critical 
work 
of 
Dr. 
Chris 
McGowan 
and 
Dr. 
Jon 
Barlow 
of 
the 
Royal 
Ontario 
Museum, 
Dr. 
David 
Gillette 
of 
New 
Mexico 
Museum 
of 
Natural 
History 
and 
Dr. 
Clayton 
Ray 
of 
the 
U.S. 
National 
Museum. 


I 
am 
also 
most 
grateful 
to 
the 
following 
Royal 
Ontario 
Museum 
staff 
for 
their 
patient 
and 
expert 
contributions: 
Pam 
Purves, 
word 
processing; 
Brian 
Iwama, 
replication 
and 
specimen 
management: 
Wasyl 
Sydorenko, 
data 
collator 
and 
bibliographer; 
and 
Martin 
Tielli, 
artist. 
The 
research 
and 
publication 
were 
made 
possible 
by 
grant 
A 
7973 
from 
the 
Natural 
Sciences 
and 
Engineering 
Research 
Council 
of 
Canada. 


ABBREVIATIONS 
AND 
ACRONYMS 
AMNH 
American 
Museum 
of 
Natural 
History, 
New 
York 
FSM 
Florida 
State 
Museum. 
University 
of 
Florida, 
Gainesville, 
Fla. 
IRV 
Irvingtonian 
Land 
Mammal 
Age 
O.R. 
Observed 
Range 
RHB 
Private 
Collection, 
Roy 
and 
Helen 
Burgess. 
Venice, 
Florida 
RLB 
Rancholabrean 
Land 
Mammal 
Age 
ROM 
Department 
of 
Vertebrate 
Palaeontology. 
Royal 
Ontario 
Museum. 
Toronto 
ROM:B 
Brayfield 
Collection, 
Royal 
Ontario 
Museum 


For 
data 
on 
collecting 
sites, 
refer 
to 
Webb 
1974 
or 
Kurten 
and 
Anderson 
1980. 


KEY 
WORDS 


Pampathere, 
Armadillo. 
Holmesina, 
Chlamytherium, 
Pampatherium,
Kraglievichia, 
Edentate, 
Xenarthra, 
Evolution. 



INTRODUCTION 


Robertson 
(1976) 
described 
a 
small 
pampathere 
of 
late 
Blancan 
age 
from 
the 
Haile 
XV 
A 
site 
in 
Florida. 
He 
recognized 
that 
it 
was 
the 
oldest 
pampathere 
yet 
described 
in 
North 
America, 
and 
called 
attention 
to 
the 
fact 
that 
there 
is 
an 
adequate 
series 
of 
specimens 
covering 
the 
time 
period 
between 
the 
Blancan 
and 
late 
Rancholabrean. 
Although 
the 
Haile 
XV 
A 
material 
includes 
pieces 
of 
skull 
and 
mandible, 
as 
well 
as 
post-cranial 
elements, 
the 
diagnosticanterior 
teeth 
or 
alveoli 
are 
lacking 
except 
for 
two 
isolated 
lower 
teeth. 
Robertson 
also 
described 
additional 
elements 
from 
two 
Blancan 
sites 
on 
the 
Santa 
Fe 
River 
to 
supplement 
his 
description.

On 
the 
basis 
of 
size, 
general 
morphology 
and 
geological 
age, 
Robertson 
referred 
his 
pampathere 
to 
the 
Argentinian 
genus 
Kraglievichia 
Castellanos 


1927, 
with 
the 
new 
species 
name 
floridanus. 
He 
also 
reviewed 
the 
taxonomyof 
the 
pampatheres 
and 
concluded 
that 
Pampatherium 
Ameghino 
1875 
was 
the 
correct 
name 
for 
the 
late 
Pleistocene 
form 
in 
North 
America 
rather 
than 
Holmesina 
Simpson 
1930. 


This 
paper 
examines 
the 
mass 
of 
pampathere 
material 
from 
North 
America, 
mainly 
from 
Florida, 
to 
document 
more 
thoroughly 
the 
chronocline 
from 
the 
Blancan 
to 
the 
Rancholabrean 
form. 
It 
also 
provides 
evidence 
that 
the 
generic 
names 
Kraglievichia 
and 
Pampatherium 
are 
not 
applicable 
to 
North 
American 
species 
and 
traces 
the 
history 
of 
discovery 
and 
taxonomy 
of 
pertinent 
North 
American 
forms. 
All 
specimens 
known 
from 
North 
and 
Central 
America 
belong 
to 
the 
Holmesina 
lineage 
with 
the 
exception 
of 
ma


terial 
from 
three 
Pleistocene 
sites 
in 
Mexico. 


MATERIAL 


The 
first 
adequate 
pampathere 
material 
of 
Blancan 
age 
from 
North 
America 
is 
that 
from 
the 
Haile 
XV 
A 
site, 
Alachua 
Co., 
Florida, 
well 
described 
byRobertson 
(1976). 
The 
holotype, 
FSM 
10902 
is 
a 
very 
fragmentary 
skull. 
Most 
of 
the 
other 
Haile 
XV 
A 
pampathere 
bones 
were 
originally 
numbered 
FSM 
10902, 
the 
same 
as 
the 
type 
skull, 
but 
bacause 
elements 
of 
more 
than 
one 
individual 
(probably 
three) 
are 
present, 
these 
have 
been 
given 
separatenumbers, 
with 
FSM 
10902 
retained 
for 
the 
skull. 
Judging 
from 
the 
state 
of 
preservation 
and 
various 
matching 
articulations, 
I 
agree 
with 
Robertson 
(op. 
cit. 
p. 
124) 
that 
many 
of 
the 
bones 
probably 
belong 
to 
a 
single 
individual. 
Robertson 
listed 
the 
bones 
from 
other 
Blancan 
sites 
as 
Referred 
Material, 
and 
more 
has 
been 
collected 
and 
identified 
in 
the 
interim. 
I 
have 
also 
examined 
and 
measured 
all 
available 
Holmesina 
material 
of 
Irvingtonian 
and 
Rancholabrean 
ages. 
The 
most 
extensive 
Irvingtonian 
collection 
(Inglis 
IA. 
Citrus 
Co.) 
is 
from 
early 
in 
the 
Irvingtonian. 
Much 
of 
the 
size 
increase 
in 
Holmesina 
apparently 
occurred 
during 
the 
Irvingtonian. 
Late 
Irvingtonian 



sites 
such 
as 
Kanopolis. 
Kansas, 
(Hibbard 
et 
al. 
1978), 
or 
McLeod, 
Florida 
(Kurten 
& 
Anderson 
1980) 
produce 
specimens 
within 
the 
expected 
size 
range 
of//. 
septentrionalis.

The 
largest 
samples, 
as 
would 
be 
expected, 
are 
Rancholabrean. 
Amongthese 
are 
FSM 
9336, 
a 
very 
well 
preserved 
individual 
lacking 
the 
skull, 
from 
Branford. 
Suwanee 
Co.; 
RHB 
3151, 
from 
Sarasota 
Co., 
less 
well 
preservedand 
also 
headless, 
but 
with 
much 
of 
the 
skeleton 
and 
carapace; 
and 
numerous 
suites 
of 
specimens 
from 
other 
localities 
in 
Florida. 
One 
of 
the 
best 
described 
specimens 
is 
that 
of 
James 
(1957) 
from 
Houston, 
Texas, 
and 
there 
are 
many 
excellent 
elements 
at 
Austin, 
Texas, 
including 
those 
from 
Ingleside, 
San 
Patricio 
Co., 
described 
by 
Lundelius 
(1972). 
Specimens 
from 
outside 
Florida, 
apparently 
all 
Rancholabrean, 
will 
be 
described 
and 
comparedin 
a 
subsequent 
paper. 
Certain 
specimens 
from 
Texas 
are 
particularly 
interestingbecauseoftheirresemblancetoH. 
humboldtifrom 
Brasil. 


HISTORY 
OF 
NORTH 
AMERICAN 
COLLECTING 


The 
first 
North 
American 
pampathere 
specimens 
were 
osteoderms 
collected 
near 
Arcadia 
on 
Peace 
Creek 
(now 
Peace 
River. 
De 
Soto 
Co.) 
in 
Florida 
in 
1888. 
Leidy 
(1889a) 
first 
described 
them 
as 
Glyptodon 
septentrionalis 
but 
later 
(1889b) 
referred 
them 
to 
the 
Brasilian 
species 
Chlamydotherium 
humboldti 
Lund 
1839. 
Sellards 
(1915a, 
1915b) 
described 
further 
material 
from 
other 
parts 
of 
Florida, 
including 
most 
of 
a 
right 
mandible 
and 
many 
osteoderms. 
He 
recognized 
these 
as 
being 
different 
from 
C. 
humboldti 
and 
revived 
Leidy’s 
specific 
name 
in 
the 
combination 
Chlamytherium 
septentrionalis 
(see 
taxonomic 
section 
below). 
In 
1922 
Cahn 
reported 
an 
excellent 
mandible 
from 
Brazos 
Co., 
Texas, 
which 
he 
identified 
as 
Chlamytherium 
septentrionalis. 
The 
original 
specimen 
was 
deposited 
in 
the 
museum 
of 
the 
Texas 
Agricultural 
and 
Mechanical 
College, 
but 
could 
not 
be 
located. 
Good 
casts 
exist 
in 
AMNH 
(No. 
13221) 
and 
ROM 
(No. 
6466). 
The 
presently 
valid 
generic 
name 
Holmesina 
was 
erected 
by 
Simpson 
(1930), 
the 
descriptionbeing 
based 
on 
the 
premaxilla, 
maxilla, 
mandible, 
and 
associated 
osteoderms 
of 
a 
single 
individual 
(AMNH 
26856) 
from 
Manatee 
Co., 
Florida, 
which 
Simpson 
designated 
the 
neotype.

In 
subsequent 
years 
numerous 
remains 
of 
pampatheres 
continued 
to 
be 
collected. 
A 
large 
proportion 
of 
these 
are 
in 
the 
Florida 
State 
Museum, 
but 
there 
is 
also 
fine 
material 
at 
Austin, 
Texas; 
Ann 
Arbor, 
Michigan; 
Toronto,

Canada; 
Los 
Angeles 
and 
Berkeley, 
California; 
and 
in 
several 
private 
collections 
in 
Florida. 
Included 
in 
these 
are 
numerous 
specimens 
dating 
from 
late 
Blancan 
and 
Irvingtonian 
as 
well 
as 
the 
much 
more 
prolific 
Rancholabrean. 


This 
material 
is 
now 
adequate 
to 
permit 
a 
review 
of 
the 
North 
American 
species, 
and 
to 
document 
their 
history 
since 
their 
first 
appearance 
in 
the 
Blancan. 



TAXONOMIC 
HISTORY 
PERTINENT 
TO 
NORTH 
AMERICA 


The 
nomenclature 
of 
the 
fossil 
giant 
armadillos 
has 
had 
a 
long, 
confused 
history. 
In 
1839 
the 
name 
Chlamytherium 
humboldti 
Lund 
was 
publishedfor 
material 
from 
Pleistocene 
cave 
deposits 
in 
the 
State 
of 
Minas-Gerais, 
Brasil. 
Lund 
hadintendedthegenericnametobeChlamydotherium 
(PaulaCouto 
1954, 
1956). 
Lund's 
manuscript, 
dated 
November 
16, 
1837 
contained 
the 
correct 
spelling, 
Chlamy 
dotherium 
and 
thus 
the 
published 
version 
was 
a 
typographical 
error. 
Lund 
subsequently 
used 
the 
correct 
spelling, 
as 
did 
many 
later 
authors 
(see 
Paula 
Couto 
1956, 
pp. 
13-14). 
To 
complicate 
matters 
further, 
it 
was 
discovered 
that 
Broun 
(1838) 
had 
proposed 
Chlamydotherium 
for 
a 
glyptodont, 
making 
the 
name 
unavailable 
for 
the 
fossil 
giantarmadillo. 
Ameghino 
(1875) 
realized 
this 
situation 
and 
erected 
the 
genusPampatherium 
to 
include 
a 
new 
species 
from 
Argentina 
(P. 
typum) 
as 
well 
as 
Lund’s 
species 
humboldti. 
Ameghino 
later 
dropped 
the 
generic 
name 
Pampatherium 
except 
when 
used 
for 
P. 
typum. 
In 
1883 
he 
briefly 
described 
the 
early 
Pliocene 
species 
C. 
paranense 
in 
the 
genusChlamydotherium,and 
subsequently 
(1888) 
C. 
intermedium 
in 
the 
same 
genus. 
Moreno 
and 
Mercerat 
(1891) 
described 
the 
oldest 
pampathere 
then 
known 
(Hayquerian, 
late 
Miocene) 
as 
Chlamydotherium 
minutum. 
Further 
material 
was 
assigned 
to 
the 
same 
species 
by 
Rovereto 
(1914). 
The 
same 
author 
(op. 
cit. 
p. 
223) 
also 
erected 
a 
new 
species, 
Chlamydotherium 
subintermedium 
on 
the 
basis 
of 
three 
isolated 
osteoderms. 


The 
most 
recent 
catalog 
of 
dasypodids 
of 
Argentina, 
(Scillato 
Yane 


1980) 
followed 
Paula 
Couto 
(1954) 
by 
retaining 
Pampatherium, 
with 
P. 
typum 
as 
the 
genotype 
of 
the 
Pampatheriinae. 
It 
did 
not 
recognize 
ChlamytheriumorChlamydotherium. 
1agreewiththerejectionofthesetwonames, 
and 
any 
derivatives 
thereof. 
The 
valid 
type 
genus 
is 
Pampatherium 
Ameghino 
1875. 


Moreno 
and 
Mercerat 
(1891) 
recognized 
the 
family 
Chlamydotheriidae,
and 
Rovereto 
(1914) 
placed 
his 
C. 
minutum 
in 
the 
Chlamydotheriidae.
Lydekker 
(1894) 
recognized 
the 
Chlamydotheriinae 
as 
a 
subfamily 
of 
the 
Dasypodidae, 
ignoring 
Ameghino’s 
(1875) 
proposal 
to 
use 
Pampatherium 
as 
the 
type 
genus 
for 
both 
the 
Brasilian 
and 
Argentinian 
late 
Pleistocene 
speciesand 
his 
recognition 
in 
1889 
of 
the 
family 
Chlamydotheriidae.

The 
group 
was 
given 
superfamilial 
status 
when 
Bordas 
(1939a) 
recognized 
its 
conservatism 
and 
longevity, 
and 
erected 
the 
superfamily 
Chlamydotherioidea. 
Hoffstetter 
(1969) 
however, 
followed 
Patterson 
(1967) 
in 
maintaining 
the 
Pampatheriinae 
as 
a 
subfamily.

Thus, 
there 
are 
historically 
two 
schools 
of 
thought 
regarding 
the 
correct 
familial 
or 
subfamilial 
name 
of 
the 
group. 
The 
weight 
of 
evidence 
would 
appear 
to 
leave 
no 
question 
of 
the 
validity 
of 
Pampatherium 
as 
the 
type 
genus 
of 
the 
subfamily 
Pampatheriinae. 
With 
the 
exception 
of 
Castellanos, 



no 
student 
familiar 
with 
both 
the 
North 
and 
South 
American 
material 
continued 
to 
use 
Chlamytherium 
or 
Chlamydotherium, 
nor 
the 
common 
name 
chlamythere. 
References 
for 
this 
position 
include 
Romer 
(1966), 
Hoffstetter 
(1969), 
Patterson 
and 
Pascual 
(1972), 
Marshall, 
Hoffstetter 
and 
Pascual 
(1983) 
and 
Cartelle 
and 
Bohorquez 
(1985).

Unfortunately, 
this 
acceptance 
has 
not 
been 
universal. 
James 
(1957)
described 
a 
pampathere 
from 
Texas, 
and 
recognized 
Chlamytherium 
as 
valid 
in 
the 
late 
Pleistocene 
of 
North 
America. 
He 
further 
suggested 
that 
the 
family 
Chlamytheriidae 
(=Chlamydotheriidae 
Ameghino 
1894) 
might 
be 
well 
founded. 
Lundelius 
(1972) 
followed 
James 
in 
using 
Chlamytherium 
septeutrionalis, 
but 
assigned 
it 
to 
the 
family 
Dasypodidae. 
Robertson 
(1976)
agreed 
with 
James 
in 
rejecting 
Holmesina 
and 
accepted 
Pampatherium 
as 
the 
valid 
genus 
to 
include 
the 
species 
septentrionalis. 
He 
also 
placed 
Pampatherium 
in 
the 
family 
Dasypodidae 
without 
designation 
of 
a 
subfamily, 
but 
refers 
to 
various 
species 
of 
fossil 
giant 
armadillos 
as 
chlamytheres.

In 
my 
description 
of 
pampathere 
armor 
(Edmund 
1985a), 
I 
considered 
the 
giant 
armadillos 
to 
be 
a 
distinct 
family, 
the 
Pampatheriidae. 
Paula 
Couto. 
in 
an 
uncompleted 
manuscript 
given 
to 
me 
in 
1981. 
recognized 
the 
family 
Pampatheriidae 
(Paula 
Couto 
1958) 
but 
the 
manuscript 
lacks 
a 
bibliography. 
In 
his 
Paliomastozoologia 
(1979) 
he 
describes 
the 
subfamily 
Pampatheriinae 
Paula 
Couto 
1958, 
but 
without 
a 
citation. 
The 
1981 
assignment 
to 
familial 
rank 
may 
be 
a 
lapsus, 
but 
he 
clearly 
lists 
it 
under 
the 
SuperfamilyDasypodoidea 
Simpson 
1931. 
I 
agree 
with 
Paula 
Couto’s 
1981 
opinion, 
and 
recognize 
the 
fossil 
giant 
armadillos 
as 
a 
separate 
family. 
From 
their 
earliest 
appearances, 
Friasian 
of 
Colombia 
and 
Chasicoan 
of 
Argentina,
pampatheres 
have 
been 
quite 
distinct 
from 
glyptodonts 
and 
dasypodids, 
and 
have 
maintained 
a 
conservative 
suite 
of 
distinguishing 
characters 
as 
follows: 


—Armadillos 
of 
large 
size 
(up 
to 
3 
m 
in 
length) 
with 
a 
robust 
skeleton. 
Manus 
and 
pes 
digitigrade, 
without 
extreme 
fossorial 
adaptations.
—Carapace 
divided 
by 
three 
transverse 
bands 
of 
imbricating 
osteoderms. 
All 
carapacial 
osteoderms 
relatively 
large, 
each 
bearing 
in 
life 
a 
single
keratinous 
scute. 
—Caudal 
armor 
typically 
one 
ring 
of 
osteoderms 
per 
vertebral 
segment.
-Mandible 
robust, 
deep 
dorsoventrally, 
with 
the 
coronoid 
process 
and 
condyle 
raised 
well 
above 
the 
occlusal 
level. 
Articular 
surfaces 
of 
condyle 
and 
glenoid 
essentially 
Hat. 
-Dental 
formula 
9/9, 
with 
the 
most 
anterior 
teeth 
cylindrical 
and 
at 
least 
the 
posterior 
five 
bilobate 
in 
section. 
Occlusal 
surfaces 
Hat, 
often 
with 
transverse 
striae, 
indicating 
herbivorous 
diet. 
All 
teeth 
permanently
open-rooted.
-Premaxilla 
bears 
one 
tooth, 
but 
with 
the 
suture 
for 
the 
maxilla 
passing
through 
the 
second 
alveolus. 



—Skull 
pneumatic, 
with 
a 
series 
of 
large 
sinuses 
in 
the 
maxilla, 
frontal 


and 
alisphenoid.

Jugal 
well 
developed, 
with 
a 
suborbital 
flange 
for 
the 
masseter 
muscle. 


Similarly 
the 
prominent 
pterygoid 
processes 
provide 
strong 
attachments 


for 
the 
pterygoid 
muscles. 


—Cervical 
vertebrae 
II 
to 
V 
coossified, 
as 
well 
as 
Cervical 
VII 
and 
Thora


cics 
I 
and 
11. 


Pelvis 
massive, 
the 
posterior 
part 
formed 
by 
greatly 
elongated 
and 
flat


tened 
transverse 
processes 
of 
at 
least 
three, 
and 
perhaps 
as 
many 
as 
five 


or 
six 
caudal 
vertebrae 
fused 
to 
the 
ischium. 
A 
broad, 
laterally 
com


pressed 
process 
descends 
to 
meet 
the 
pubis, 
enclosing 
the 
large 
sacro


ischiac 
fenestra, 
larger 
than 
the 
obturator 
foramen. 
The 
synsacrum 
con


sists 
of 
approximately 
nine 
vertebrae. 


Some 
of 
these 
characters 
are 
found 
also 
in 
the 
Dasypodidae 
s.s. 
and 
Glyptodontidae, 
but 
the 
pampatheres 
have 
an 
adequate 
suite 
of 
distinctive 
characters 
to 
merit 
recognition 
as 
a 
family.

Alfredo 
Castellanos, 
the 
foremost 
student 
of 
Argentinian 
pampatheres,
(1927, 
1937, 
1946, 
1958) 
followed 
Ameghino’s 
abandonment 
of 
the 
genusPampatherium 
and 
the 
derived 
subfamilial 
name. 
He 
retained 
the 
species 
typum 
and 
humboldti 
but 
placed 
them 
first 
(1927) 
in 
the 
genus 
Chlamydotherium 
and 
subfamily 
Chlamydotheriinae. 
In 
his 
subsequent 
publicationshe 
used 
Chlamytherium. 
Chlamytheriinae 
and 
the 
common 
name 
clamiterios. 


SUMMARY 
OF 
PAMPATHERE 
SYSTEMATICS 


The 
main 
thesis 
of 
this 
paper 
is 
the 
documentationof 
a 
chronocline 
leadingfrom 
the 
Blancan 
H. 
floridanus 
to 
the 
Rancholabrean 
H. 
septentrionalis.
However, 
because 
of 
the 
confusing 
plethora 
of 
inappropriate 
generic 
names 
in 
the 
recent 
North 
American 
literature 
(James 
1957, 
Lundelius 
1972, 
Robertson 
1978) 
a 
summary 
of 
pampathere 
systematics, 
as 
I 
currently 
perceive 
it, 
is 
pertinent.

Scillato 
Yane 
(1980) 
conveniently 
listed 
all 
genera 
and 
species 
recognized 
from 
Argentina, 
along 
with 
an 
exhaustive 
synonymy. 
Literature 
on 
taxa 
from 
other 
South 
American 
countries 
is 
scattered, 
and 
the 
entire 
grouprequires 
thorough 
revision. 
Castellanos 
(1927. 
1937, 
1946) 
proposed 
a 
lineage 
of 
pampatheres 
increasing 
in 
size 
and 
in 
degree 
of 
anterior 
“molarization”, 
covering 
the 
time 
period 
from 
the 
Chasicoan 
(Upper 
Miocene) 
to 
the 
Lujanian 
(Upper 
Pleistocene). 
Three 
generic 
names, 
Vassallia, 
Plaina, 
and 
Pampatherium 
were 
assigned 
to 
parts 
of 
this 
apparently 
continuous 
lineage.
Kraglievichia 
paranense 
was 
erected 
by 
Castellanos 
(1927, 
1937) 
for 
a 
small 
pampathere 
typified 
by 
peculiarly 
ornamented 
osteoderms, 
and 
apparentlyranging 
from 
the 
Chasicoan 
to 
the 
Montehermosan 
(Upper 
Miocene 
to 



8 


Lower 
Pliocene). 
It 
may 
be 
closely 
related 
to 
Vassallia 
but 
the 
material 
re


covered 
to 
date 
is 
inadequate 
for 
a 
decision. 


Another 
taxon, 
as 
yet 
unnamed, 
is 
known 
from 
the 
Montehermosan 
ot 
Argentina, 
and 
is 
represented 
only 
by 
numerous 
isolated 
small 
osteoderms 
with 
clearly 
demarcated 
submarginal 
bands 
and 
narrow 
central 
keels. 
Verysimilar, 
if 
not 
identical 
osteoderms 
were 
collected 
on 
the 
Rio 
Jurua 
in 
the 
state 
of 
Acre 
in 
Brasil 
in 
1956 
by 
G.G. 
Simpson 
and 
Carlos 
de 
Paula 
Couto 
(unpublished).

Similar 
but 
not 
identical 
osteoderms 
are 
associated 
with 
cranial 
and 
ske


letal 
elements 
from 
the 
Friasian 
La 
Venta 
locality 
of 
Colombia 
(Marshall,

Hoffstetter 
and 
Pascual 
1983). 
Isolated 
osteoderms 
from 
this 
site 
were 


referred 
to 
Kraglievichia 
by 
Porta 
(1962), 
but 
this 
was 
an 
error 
in 
identifi


cation. 
The 
La 
Venta 
material 
represents 
the 
oldest 
known 
typical 
pampa


there, 
since 
the 
status 
of 
the 
Eocene 
genus 
Machlydotherium 
is 
still 
unclear. 


The 
Brasilian 
species 
Holmesina 
humboldti 
(Lund 
1839), 
along 
with 


several 
synonymic 
species 
names, 
have 
been 
referred 
to 
various 
genera, 
most 


often 
to 
Pampatherium 
(Cartelle 
and 
Bohorquez 
1985). 
1 
believe 
it 
is 
more 


closely 
related 
to 
Holmesina 
and 
refer 
it 
to 
that 
genus 
pending 
more 
detailed 


study.

Also 
probably 
referable 
to 
Holmesina 
is 
the 
species 
H. 
occidentale 
Hoffstetter 
1952, 
referred 
to 
Chlamytherium 
by 
Hoffstetter 
(1952), 
to 
Holmesina 
by 
Hoffstetter 
(1953) 
and 
to 
Hoffstetteria 
by 
Castellanos 
(1957). 
I 
have 
collected 
a 
large 
series 
of 
this 
species 
from 
Peru 
and 
Ecuador 
and 
have 
also 
recognized 
identical 
specimens 
from 
western 
Venezuela. 
Cartelle 
and 
Bohorquez 
(1985) 
described 
Pampatherium 
paulacoutoi 
from 
Bahia, 
Brasil. 
Their 
measurements 
and 
descriptions 
agree 
closely 
with 
the 
occidentale 
material 
at 
the 
Royal 
Ontario 
Museum 
and 
may 
prove 
synonymous. 
1 
agreewith 
Hoffstetter 
that 
Holmesina 
is 
probably 
the 
most 
appropriate 
genus 
for 
occidentale, 
and 
that 
it 
is 
not 
closely 
related 
to 
the 
Pampatherium 
lineage.
Unfortunately, 
Pliocene 
and 
early 
Pleistocene 
sites 
are 
essentially 
unknown 
in 
Central 
America 
and 
the 
northern 
part 
of 
South 
America, 
making 
it 
difficult 
to 
do 
more 
than 
speculate 
on 
the 
history 
and 
relationship 
of 
Holmesina 
in 
Florida 
and 
in 
South 
America 
(Robertson 
1976, 
p. 
142). 



INCREASING 
SIZE 
OF 
HOLMESINA 
IN 
FLORIDA 


Robertson 
(1976) 
described 
pampathere 
material 
from 
the 
late 
Blancan 
site 
Haile 
XV 
A, 
and 
compared 
various 
elements 
to 
those 
from 
Irvingtonian 
and 
Rancholabrean 
sites 
in 
Florida. 
He 
clearly 
demonstrated 
a 
large 
increase 
in 
size 
during 
this 
period 
of 
approximately 
two 
million 
years 
(Savage 
and 
Russel 
1983). 
In 
addition 
to 
the 
elements 
figured 
by 
Robertson, 
osteoderms 
and 
other 
bones 
have 
been 
measured 
wherever 
adequate 
material 
exists 
for 
comparison. 
Unfortunately, 
elements 
comparable 
at 
intervals 
during 
the 
Blancan, 
Irvingtonian 
and 
Rancholabrean 
are 
few, 
so 
that 
statistical 
treatment 
is 
not 
possible. 
The 
Irvingtonian, 
in 
particular, 
is 
represented 
mainlyby 
one 
very 
rich 
but 
early 
site 
(Inglis 
I 
A) 
and 
by 
a 
single 
very 
late 
individual 
(the 
McLeod 
site). 
Thus, 
in 
many 
cases 
the 
comparison 
can 
be 
made 
between 
only 
Blancan 
and 
Rancholabrean. 
In 
this 
section, 
all 
material 
is 
from 


Uorida. 


CARAPACE 


The 
morphology 
of 
typical 
buckler 
and 
imbricating 
osteoderms 
is 
given 
in 
Edmund 
1 
985a. 
In 
both 
types 
of 
osteoderms 
a 
depressed 
marginal 
band 
extends 
completely 
around 
the 
buckler 
osteoderms, 
and 
around 
much 
of 
the 
ornamented 
area 
of 
the 
imbricating 
osteoderms. 
It 
is 
generally 
one 
to 
three 
millimeters 
wide, 
and 
is 
rugose 
and 
pitted 
(Fig. 
1A). 
Concentric 
within 
this 
marginal 
band 
is 
the 
submarginal 
band, 
usually 
prominently 
raised 
above 
the 
marginal 
band 
and 
the 
depressed 
central 
area. 
It 
varies 
in 
height 
and 
width, 
but 
is 
generally 
narrow 
(2 
4 
mm) 
and 
clearly 
elevated. 
The 
raised 
central

-

figure 
is 
variable, 
but 
is 
separated 
from 
the 
submarginal 
band 
by 
a 
depressed,
usually 
finely 
punctate 
area 
termed 
the 
sulcus. 


In 
Holmesina 
septentrionalis, 
especially 
in 
large, 
well-preserved 
individuals, 
the 
features 
described 
above 
are 
usually 
strongly 
expressed. 
In 
the 
early 
Irvingtonian 
and 
Blancan 
specimens, 
the 
relief 
of 
the 
surface 
features 
is 
less 
pronounced, 
but 
still 
clear 
(Fig. 
IB). 
This 
ornamentation 
separates 
the 
Holmesina 
group 
from 
the 
feebly 
ornamented 
Vassallia 
Plaiua 
Pampatherium 
group 
. 
In 
the 
latter, 
the 
marginal 
depressed 
band 
is 
usually 
absent 
or 
variable, 
and 
the 
depressed 
area 
between 
the 
submarginal 
band 
and 
central 
figure 
is 
absent 
or 
very 
feebly 
expressed 
(Edmund 
1985b). 
Kraglievichia 
has 
an 
ornamentation 
that 
apparently 
is 
unique 
(Fig. 
1C), 
although 
definitely

associated 
skeletal 
and 
carapacial 
material 
is 
unknown. 
The 
remainder 
of 
this 
section 
wall 
deal 
with 
Holmesina 
osteoderms 
from 
Florida 
only, 
so 
as 
to 
limit 
the 
variability 
attributable 
to 
geographic 
variation. 


It 
is 
assumed 
that 
the 
number 
of 
osteoderms 
in 
each 
carapace 
is 
nearly 
constant,and 
thatan 
adequate 
sample 
ofosteodermsreflects 
thesizeofthe 
individual. 
I 
make 
this 
assumption 
mainly 
on 
the 
basis 
that 
small 
(Blancan)
specimens 
have 
small 
osteoderms, 
but 
that 
the 
shapes 
and 
proportions 
are 



Table 
1. 
Increase 
in 
osteoderm 
size 
between 
Blancan 
and 
Rancholabrean 
samples. 


Type 
Land 
Mammal 
Age 
O.R. 
Area 
N 
X 
Buckler 
Blancan 
678-751 
40 
714 
RLB 
1368-1970 
88 
1628 
Imbricating 
Blancan 
RLB 
1675-1802 
3399-4148 
20 
23 
1738 
3847 


=

Increment 
ratio 
(Buckler) 
1628/7142.28 
Increment 
ratio 
(Imbricating) 
3847/1738 
= 
2.21 


Fig. 
1. 
Two 
buckler 
and 
one 
imbricating 
osteoderms 
ofeach 
of 
three 
species 
of 
pampatheres. 
A) 
ROM:B 
7960 
Holmesina 
floridanus, 
Blancan,
El 
Jobean, 
Charlotte 
Co., 
Florida. 
B) 
FSM 
9336 
Holmesina 
septentrionalis, 
Rancholabrean, 
Branford. 
Suwanee 
Co., 
Florida. 
C) 
ROM 
23829 


Kraglievichia 
paranense, 
Montehermosan, 
Entre 
Rios, 
Argentina. 



similar 
to 
those 
of 
larger, 
Rancholabrean 
individuals. 
Further, 
the 
genus 
is 
exceptionally 
conservative. 
Unfortunately 
nothing 
has 
been 
reported 
in 
the 
literature 
regarding 
variability 
in 
number 
of 
osteoderms, 
either 
in 
modern 
armadillos 
or 
in 
the 
fossil 
record. 
Table 
1 
displays 
the 
“area” 
—i.e. 
lengthtimes 
width, 
for 
both 
imbricating 
and 
buckler 
osteoderms, 
plotted 
againsttime 
as 
represented 
by 
land 
mammal 
ages. 
Each 
sample 
represents 
the 
mean 
of 
at 
least 
ten 
osteoderms 
from 
a 
single 
individual 
or 
a 
single 
site. 
Between 
the 
Blancan 
and 
Rancholabrean, 
the 
area 
of 
both 
the 
buckler 
and 
imbricating 
osteoderms 
increased 
by 
a 
factor 
of2.2, 
suggesting 
that 
the 
body 
surface 
area 
increased 
approximately 
2.2 
times. 
This 
should 
correlate 
with 
the 
increasing 
size 
of 
the 
skeleton, 
using 
the 
formula 
A 
L 
2 
A, 
where 
Lis 
the 
length 
of 
the 
skeletal 
element 
and 
A 
is 
the 
surface 
area. 
The 
mean 
for 
the 
squares 
of 
limb 
bone 
length 
(Table 
2) 
increase 
is 
2:26, 
virtually 
equal 
to 
the 


increase 
in 
osteoderm 
area. 
This 
is 
one 
of 
the 
few 
instances 
in 
which 
this 
relationship 
can 
be 
confirmed 
for 
vertebrate 
fossil 
material. 


LIMB 
BONES 


Robertson 
(1976) 
compared 
the 
femora 
and 
calcanea 
of 
Florida 
Holmesina 
of 
various 
land 
mammal 
ages 
by 
photographs 
in 
his 
figures 
8 
and 
9, 
but 
did 
not 
compare 
measurements. 
Comparable 
elements 
are 
not 
always 
available 
for 
all 
three 
land 
mammal 
ages, 
but 
some 
can 
be 
obtained 
over 
the 
entire 
period 
(Table 
2). 
During 
the 
time 
from 
Blancan 
to 
Rancholabrean, 
the 
lengths 
of 
the 
radius, 
tibia-fibula 
and 
calcaneum 
increased 
by 
approximately 
one 
and 
a 
half 
times 
(Table 
2). 
As 
noted 
by 
Robertson 
(op. 
cit.), 
there 
is 
no 
obvious 
change 
in 
morphology 
of 
the 
tibia-fibula 
or 
the 
radius. 
Although 
one 
might 
expect 
the 
Rancholabrean 
animals 
to 
have 
stouter 
long 
bones 
than 
theBlancananimals,thisisnotalwaysthecase. 
Theratiosofdistalwidthto 
total 
length 
of 
the 
tibia-fibula 
are 
shown 
in 
Table 
3. 
The 
Rancholabrean 
sample 
appears 
to 
be 
ten 
percent 
more 
robust 
than 
the 
Blancan. 
but 
the 
sample 
size, 
especially 
in 
the 
Blancan, 
is 
small. 


Table 
2. 
Increase 
in 
bone 
lengths 
from 
Blancan 
to 
Rancholabrean 
times. 


Element 
Blancan 
Rancholabrean 
Ratio 
of 
Ratio 
Mean 
Length 
Mean 
Length 
RLB 
length 
to 
squaredBlancan 
length 


Radius 
90 
133 
1.48 
2.18 


Tibia-fibula 
136 
195 
1.43 
2.06 


Calcaneum 
64 
102 
1.58 
2.54 


Means 
1.50 
2.26 



Table 
3.Changes 
intibia-fibulameasurementsfrom 
Blancanto 
Rancholabreantimes. 


Land 
Mammal 
N 
Distal 
Width 
Length 
Ratio 
of 
Age 
Distal 
W. 
to 
Length 
x7

O.R. 
X 
O.R. 
RLB 
6 
78-93 
84 
185-213 
195 
0.43 
Blancan 
6 
47-60 
54 
122-151 
136 
0.39 
Table4.Changesinradius 
dimensionsfromBlancantoRancholabreantimes. 


Land 
Mammal 
N 
Distal 
Width 
Length 
Ratio 
of 
Age 
Distal 
W. 
to 
Length

O.R. 
X 
O.R. 
>T 
RLE 
2 
41-41 
41 
131-135 
133 
0.31 
*

IRV 
i36-36 
36 
133—133* 
133 
0.25 
Blancan 
3 
19-22 
21 
88-94 
90 
0.23 


*left 
and 
right 
of 
same 
individual 
(Yarmouthian) 


Table 
5.Changes 
in 
calcaneumdimensionsfromBlancan 
toRancholabreantimes. 


Land 
Mammal 
N 
Tuber 
Width 
Length 
Ratio 
of 
Age 
Tuber 
W. 
to 
Length

O.R. 
X 
O.R. 
X 
RLB 
4 
46-52 
49 
95-107 
102 
0.48 
IRV 
4 
35-40 
37 
80-102 
89 
0.42 
Blancan 
4 
26-26 
26 
63-64 
64 
0.41 
Because 
of 
inadequate 
samples, 
the 
only 
other 
long 
bone 
available 
on 


which 
to 
check 
this 
trend 
is 
the 
radius 
(Table 
4). 
Distal 
width 
was 
divided 


by 
total 
bone 
length, 
indicating 
an 
increase 
in 
robustness 
over 
the 
three 
land 
mammal 
ages. 
The 
percentage 
increase 
from 
Blancan 
to 
Rancholabrean 
is 
30 
percent. 
This 
is 
based 
on 
a 
total 
of 
only 
seven 
specimens.

A 
larger 
sample 
is 
available 
for 
the 
calcaneum 
(Table 
5), 
showing 
that 
it 
increases 
both 
in 
total 
length 
and 
width 
of 
tuber 
calcanei. 
The 
tuber 
be



Table6.Changesintheratiooftubertoneckin 
thecalcaneumfromBlancantoRan


cholabrean 
times. 
Land 
Mammal 
N 
Tuber 
Width 
Neck 
Thickness 
Ratio 
of 
Age 
Tuber 
W 
to 
Neck 
Thickness 
O.R. 
x 
O.R. 
X 
RLB 
5 
46-52 
49 
16-19 
17 
2.7 
IRV 
4 
35-40 
37 
14-17 
15 
2.5 
Blancan 
2 
26 
26 
11-12 
11.5 
2.3 


Table7. 
Changesintheseparationofmedialandlateralfacetsonthecalcaneumand 
astragalus 
from 
Blancan 
to 
Rancholabrean 
times. 


Land 
Mammal 
Calcaneum 
AstragalusAge 


N 
O.R. 
x" 
N 
O.R. 
X 
RLB 
43-7 
6 
54-7 
6 
IRV 
3 
0-1 
0.3 
42-5 
3 
Blancan 
20 
0 
0 


._ 


Table 
8. 
Changes 
in 
dimensions 
of 
facets 
on 
the 
calcaneum 
from 
Blancan 
to 
Rancholabrean 
times. 


Land 
N 
Medial 
facet 
Lateral 
facet 
Areal 
Ratio 
of 


==

Mammal 
“Area” 
LxW 
“Area” 
LxW 
Medial 
facet 
to 
Age 
Lateral 
facet 


O.R. 
X 
O.R. 
X 
RLE 
4 
210-397 
303 
384-696 
516 
0.59 
IRV 
4 
195-266 
238 
320-468 
401 
0.59 
Blancan 
4 
117 
117 
204 
204 
0.57 
comes 
more 
prominent 
and 
rounded, 
and 
increases 
in 
width 
faster 
than 
length 
(Fig. 
2 
and 
Table 
5). 
Another 
change, 
noted 
by 
Robertson, 
is 
that 
the 
tuber 
becomes 
expanded 
relative 
to 
the 
post-articular 
segment. 
This 
is 
shown 
in 
Table 
6, 
in 
which 
the 
tuber 
width 
is 
divided 
by 
minimum 
thickness 
of 
the 
neck. 
The 
increase 
is 
quite 
regular 
during 
the 
three 
stages, 
giving 
a 
total 
increase 
of 
16 
percent. 



The 
other 
major 
change, 
as 
noted 
by 
Robertson, 
is 
in 
the 
separation 
of 
the 
two 
facets 
for 
articulation 
between 
the 
astragalus 
and 
calcaneum 
(Fig. 
2 
and 
Table 
7). 
In 
the 
two 
available 
calcanea 
from 
the 
Blancan, 
the 
two 
facets 
are 
thoroughly 
confluent 
over 
a 
length 
of 
several 
millimeters, 
the 
distinction 
indicated 
only 
by 
a 
change 
of 
angle. 
By 
early 
Irvingtonian 
time(Inglis 
I 
A)
the 
facets 
are 
discrete, 
but 
still 
in 
contact 
in 
two 
of 
the 
three 
known 
specimens, 
and 
in 
the 
Rancholabrean 
they 
are 
always 
separate. 
No 
astragalus 
is 
known 
from 
the 
Blancan, 
but 
those 
from 
the 
Irvingtonian 
and 
Rancholabrean 
show 
considerably 
wider 
separation 
of 
facets 
than 
in 
the 
corresponding 
calcanea. 


Robertson 
also 
mentioned 
that 
the 
two 
facets 
are 
nearly 
equal 
in 
size 
in 
the 
Blancan 
form, 
but 
that 
the 
lateral 
facet 
is 
much 
larger 
in 
the 
Rancholabrean. 
In 
the 
case 
of 
the 
Rancholabrean 
specimen 
FSM 
9336, 
which 
Robertson 
had 
for 
comparison, 
the 
medial 
facet 
is 
particularly 
small. 
To 
resolve 
this 
problem 
with 
a 
larger 
sample, 
the 
length 
and 
width 
measurements 
of 
the 
facets 
of 
a 
number 
of 
specimens 
were 
compared 
(Table 
8). 
The 
difference 
appears 
minimal, 
although 
the 
number 
of 
specimens 
studied 
is 
small. 


It 
should 
be 
noted 
that 
the 
analyses 
of 
the 
calcaneum 
are 
based 
on 
fairlysmall 
sample 
sizes. 
However, 
there 
is 
no 
evidence 
to 
the 
contrary, 
and 
1 
feel 
confident 
that 
larger 
sample 
sizes 
would 
only 
give 
added 
weight 
to 
the 
conclusion 
of 
gradualism. 
The 
same 
observation 
applies 
to 
the 
study 
of 
the 
metapodials. 


Fig. 
2. 
Left 
calcanea 
of 
Floridean 
species 
of 
Holmesina. 
A) 
RHB 
3151 


H. 
septentrionalis 
Rancholabrean, 
Sarasota 
Co. 
B) 
FSM 
20954 
Holmesina 
sp., 
Irvingtonian, 
.Haile 
XVI 
A 
(reversed 
from 
right). 
C) 
FSM 
20946 
Holmesina 
sp.. 
Irvingtonian, 
Inglis 
I 
A. 
D) 
ROM 
21803 
H. 
floridanus, 
Blancan, 
Haile 
XV 
A 
(cast). 

METAPODIALS 


Metapodials, 
especially 
numbers 
II 
to 
IV. 
are 
among 
the 
most 
frequently 
preserved 
skeletal 
elements 
of 
pampatheres. 
All 
specimens 
with 
adequate 
data 
are 
plotted 
in 
Figs. 
3 
and 
4. 
which 
show 
a 
clear 
increase 
in 
size 
from 
Blancan 
to 
Rancholabrean. 
The 
mean 
increase 
in 
length 
for 
metatarsals 
11, 
111 
and 


IV. 
and 
metacarpals 
II 
and 
111 
is 
1.6 
times. 
Metacarpal 
IV. 
however, 
shows 
an 
ultimate 
increase 
of 
10 
percent. 
It 
had 
increased 
from 
28 
mm 
for 
a 
single 
Blancan 
example 
to 
a 
mean 
of 
39 
mm 
in 
the 
Irvingtonian 
(N 
= 
3), 
then 
=

diminished 
to 
a 
mean 
of 
30 
mm 
in 
the 
Rancholabrean 
(N 
3). 
This 
reversal 
may 
well 
reflect 
a 
real 
trend, 
considering 
that 
the 
range 
of 
measurements 
does 
not 
overlap 
and 
that 
the 
means 
are 
well 
separated. 
A 
larger 
sample, 
especially 
with 
late 
Irvingtonian 
specimens, 
might 
cause 
the 
length 
ranges 
in 
the 
Irvingtonian 
and 
Rancholabrean 
to 
become 
confluent, 
but 
I 
would 
not 
expect 
the 
downturn 
of 
size 
to 
be 
affected. 


The 
morphology 
of 
the 
metapodials 
is 
fairly 
constant 
during 
the 
relevant 
time 
period, 
except 
for 
the 
features 
noted 
by 
Robertson. 
The 
metapodialsof 
Irvingtonian 
specimens 
resemble 
those 
of 
the 
Rancholabrean 
more 
closelythan 
they 
do 
those 
of 
the 
Blancan. 


TAXONOMIC 
CONCLUSIONS 


Robertson 
(1976) 
erected 
the 
name 
Kraglievichia 
flohdamis 
for 
the 
Blancan 
species 
of 
pampathere 
from 
Florida, 
based 
partly 
on 
Castellanos’ 
(1927,1937) 
descriptions 
of 
K. 
paranense 
from 
the 
lower 
Pliocene 
of 
Argentina,
andpartlyonosteodermsfrom 
theLaVentaMioceneofColombiaidentified 
incorrectly 
as 
Kraglievichia 
by 
Porta 
(1962). 
Robertson 
had 
access 
to 
a 
larger 
suite 
of 
material 
from 
Colombia, 
and 
suggested 
that 
it 
might 
better 
be 
referred 
to 
Castellanos’ 
(1927) 
genus 
Vassallia. 


He 
drew 
attention 
to 
the 
fact 
that 
in 
the 
Florida 
Blancan 
specimens 
the 
two 
calcaneal 
facets 
on 
the 
astragalus 
(i.e. 
ectal 
and 
sustentacular) 
are 
confluent, 
separated 
only 
by 
a 
change 
in 
angle 
(Fig. 
2). 
1 
have 
recently 
compared 
the 
Florida 
specimen 
with 
newly 
prepared 
astragali 
and 
calcanea 
from 
La 
Venta, 
and 
the 
structure 
of 
the 
astragalo-calcaneal 
articulation 
is 
quitedifferent. 
In 
the 
La 
Venta 
animal 
the 
two 
facets 
are 
not 
only 
separated 
by 
a 


gap 
of 
some 
three 
millimeters 
(in 
both 
astragalus 
and 
calcaneum) 
but 
also 
by 
a 
difference 
in 
elevation 
of 
the 
facets. 
This 
presumably 
derived 
feature 
rules 
out 
the 
La 
Venta 
pampathere 
from 
ancestry 
of 
H. 
floridanus 
since 
it 
would 
require 
the 
facets 
to 
become 
confluent 
then 
to 
separate 
during 
the 
Plio-
Pleistocene, 
an 
unlikely 
event. 


The 
ornamentation 
pattern 
of 
the 
La 
Venta 
osteoderms 
is 
generally 
similar 
to 
that 
of 
the 
North 
American 
Holmesina 
lineage, 
however, 
suggesting 
at 
least 
some 
degree 
of 
relationship. 
The 
La 
Venta 
animal 
is 
also 
distinct 
from 
the 
Vassallia 
Plaina 
Pampatherium 
lineage, 
based 
on 
osteoderm 
ornament



ation 
and 
absence 
of 
rectangular 
osteoderms 
in 
the 
pelvic 
buckler 
(Edmund

1 
985A).

Kraglievichia 
paranense 
can 
be 
distinguished 
from 
the 
La 
Venta 
animal, 
the 
Holmesina 
lineage, 
and 
the 
Pampatherium 
lineage 
by 
its 
characteristic 
osteoderms 
(Fig. 
1C). 
Although 
many 
specimens, 
including 
a 
fine 
skull, 
are 
known 
from 
Argentina, 
in 
no 
example 
are 
the 
cranial, 
post-cranial 
and 
carapacial 
elements 
associated. 
It 
is 
unfortunate 
that 
the 
type 
skull 
ofH. 
flori


danus 
is 
so 
incomplete, 
precluding 
more 
detailed 
comparison 
with 
small 
species 
from 
South 
America. 


The 
animal 
described 
by 
Robertson 
(1976) 
is 
essentially 
a 
small 
version 
of 
Holmesina 
septentrionalis. 
The 
osteoderms 
are 
quite 
dissimilar 
to 
those 
of 
Kraglievichia 
and 
of 
the 
Pampatherium 
lineage, 
and 
the 
calcaneo-astragalar 
articulation 
rules 
out 
descent 
from 
the 
La 
Venta 
animal. 
Its 
ancestry 


Fig. 
3. 
Relation 
between 
metacarpal 
length 
and 
time 
in 
Holmesina 
from 
Florida. 



may 
lie 
with 
an 
as 
yet 
undescribed 
taxon 
with 
small, 
keeled 
osteoderms 
from 
the 
Montehermosan 
of 
Argentina. 
For 
the 
purposes 
of 
this 
paper, 
it 
is 
sufficient 
to 
conclude 
that 
Holmesina 
of 
North 
America 
does 
not 
have 
close 
relationships 
with 
any 
described 
pre-Pleistocene 
South 
American 
taxon 
or 
lineage. 
Cranial 
and 
dental 
morphology 
are 
of 
little 
use 
for 
this 
purpose 
since 
there 
is 
almost 
no 
comparable 
pre-Pleistocene 
South 
American 
material 
except 
that 
assigned 
to 
Kraglievichia, 
Vassallia 
or 
Plaina. 
Thus, 
Holmesina 
is 


the 
valid 
genus 
for 
the 
Florida 
material, 
with 
H. 
floridanus 
Robertson 
1976 
in 
the 
Blancan 
and 
H. 
septentrionalis 
(Leidy 
1889) 
Simpson 
1930 
in 
the 
Rancholabrean 
and 
probably 
also 
the 
late 
Irvingtonian. 
Specimens 
from 
early 
and 
middle 
Irvingtonian, 
intermediate 
in 
size, 
are 
referred 
only 
to 
Holmesina 
sp., 
because 
a 
new 
specific 
name 
is 
unwarranted. 


Fig. 
4. 
Relation 
between 
metatarsal 
length 
and 
time 
in 
Holmesina 
from 
Florida. 



REFERENCES 


Ameghino, 
F. 
1875. 
Notas 
sobre 
fosiles 
nuevos 
de 
la 
formacion 
Pampaeana. 
Ameghino 
Obras 
Completas, 
11:13-17. 


—. 


1883. 
Sobre 
una 
coleccion 
de 
mami'feros 
fosiles 
del 
piso 
mesopotamico 
de 
la 
formation 
patagonica, 
recogidos 
en 
las 
barrancas 
del 
Parana 
por 
el 
profesor 
Pedro 
Scalabrini. 
Bob 
Acad. 
Nac. 
Cienc. 
Cordoba, 
5:10-116. 


1888. 
Lista 
de 
las 
especies 
de 
mamiferos 
fosiles 
del 
Mioceno 
superior 
de 
Monte 
Hermoso 
hasta 
ahora 
conocidas. 
Buenos 
Aires, 
21 
pp. 


—. 
1894. 
Sur 
les 
ongules 
fossiles 
de 
PArgentine. 
(Examen 
critique 
de 
Pouvrage 
de 
M.R. 
Lydekker: 
A 
study 
of 
the 
extinct 
ungulates 
of 
Argentina). 
Ameghino 
Obras 
Completas, 
XE7-203. 


Bordas, 
A.F. 
1939. 
Craniometna 
y 
region 
auditiva 
de 
Chlarnytherium 
typumAmeghino, 
Physis, 
Buenos 
Aires,T.14,447-460. 


Broun. 
M. 
1838. 
Lethaea 
geognostica. 
Stuttgart. 
Vol. 
2:545-1350 


Calm, 
A.R. 
1922. 
Chlamytherium 
septenthonalis, 
a 
fossil 
edentate 
new 
to 
the 
fauna 
of 
Texas. 
Jour. 
Mammal, 
3(l):22-24.

Cartelle, 
C. 
and 
G.A. 
Bohorquez. 
1985. 
Pampatherium 
paulacoutui, 
una 
nova 
especie 
de 
Tatu 
gigante 
de 
Bahia, 
Brasil 
(Edentata, 
Dasypodidae).
Revista 
Brasiliera 
de 
Zoologia, 
Sao 
Paulo, 
2(4):229-254.

Castellanos, 
A. 
1927. 
Breves 
notas 
sobre 
los 
clamidoterios. 
Publ. 
del 
Centro 
Estudiantes 
de 
Ingenieria 
de 
Rosario 
(Argentina), 
1-8. 


—. 
1937. 
Anotaciones 
sobre 
las 
h'neas 
filogeneticas 
de 
los 
clamiterios. 
Publ. 
Inst. 
Fisiogr. 
y 
Geob, 
Univ. 
Nac. 
Litoral, 
Rosario 
Argentina.
Serie 
Tecnica 
y 
Cientifica, 
8:1 
-35. 


—. 
1946. 
Una 
nueva 
especie 
de 
clamiterio, 
Vassalia 
maxima 
n. 
sp.
Publ. 
Inst. 
Fisiogr. 
y 
Geob, 
Univ. 
Nac. 
Litoral, 
Rosario, 
Argentina 


—. 


26: 
1-47. 
—. 
1958. 
Nota 
preliminar 
sobre 
nuevos 
restos 
de 
mamiferos 
fosiles 
en 
el 
Brocherense 
de 
Valle 
de 
los 
Reartes 
(Provincia 
de 
Cordoba, 
Argentina). 
Publ. 
20th 
Internat. 
Geob 
Congr., 
Sec. 
VII, 
217-233. 


Edmund, 
A.G. 
1985a. 
The 
armor 
of 
fossil 
giant 
armadillos 
(Pampatheriidae,
Xenarthra, 
Mammalia). 
Pearce-Sellards 
Series, 
Texas 
Memorial 
Museum,
University 
of 
Texas 
at 
Austin, 
40:1-20. 


—. 


1985b. 
The 
fossil 
giant 
armadillos 
of 
North 
America 
(Pampatheriinae, 
Xenarthra=Edentata), 
in 
The 
Evolution 
and 
Ecology 
of 
Armadillos, 
Sloths 
and 
Vermilinguas 
(Mammalia, 
Xenarthra=Edentata). 
Teri 


L. 
Davis, 
ed. 
Smithsonian 
Instution 
Press, 
Washington, 
DC, 
400 
pp.
Hibbard, 
C.W., 
R. 
J. 
Zakrewski, 
R.E. 
Eshelman, 
G. 
Edmund, 
C.D. 
Griggsand 
C. 
Griggs. 
1978. 
Mammals 
from 
the 
Kanapolis 
local 
fauna, 
Pleistocene 
(Yarmouth) 
of 
Ellsworth 
Co., 
Kansas. 
Contr. 
Mus. 
Paleont., 
Univ. 
Michigan, 
25(2): 
11-44. 



Hoffstetter, 
R. 
1952. 
Mammiferes 
Pleistocenes 
de 
la 
Republique 
de 
PEquateur. 
Mem. 
Soc. 
Geol. 
de 
France 
(N.S.) 
66, 
391 
pp. 
—. 


1953. 
Sur 
la 
presence 
d’un 
Tatou 
geant 
du 
genre 
Holmesina 
dans 
le 
Pleistocene 
de 
FEquateur 
(Amerique 
du 
Sud). 
C.R. 
Soc. 
Geol., 
France, 
6:101-102. 


. 
1969. 
Remarques 
sur 
la 
phylogenie 
et 
la 
classification 
des 
edentes 
xenarthres 
(Mammiferes) 
actueles 
et 
fossiles. 
Bull, 
du 
Museum 
Nat. 
d'Hist. 
2e 
Serie, 
41(1);91-103.

James, 
G. 
1957. 
An 
edentate 
from 
the 
Pleistocene 
of 
Texas. 
J. 
Paleont.,
31 
(4):796-808.
Kurten, 
B. 
and 
E. 
Anderson. 
1980. 
Pleistocene 
Mammals 
ofNorth 
America. 
Columbia 
Elniversity 
Press, 
New 
York, 
442 
pp.
Leidy, 
J. 
1889a. 
Description 
of 
the 
vertebrate 
remains 
from 
Peace 
Creek,
Florida. 
Trans. 
Wagner 
Free 
Inst. 
Sci., 
2:19-32. 


—. 


1889b. 
Fossil 
vertebrates 
from 
Florida. 
Proc. 
Acad. 
Nat. 
Sci. 
Phila., 
1889:96-7. 


Fund, 
P.W. 
1839. 
Blik 
paa 
Brasiliens 
Dyreverden 
for 
Sidste 
Jordomvaeltning. 
Anden 
afhandling: 
Pattedyrene. 
Fagoa 
Santa, 
d. 
16d. 
Novbr. 
1837. 
Det 
Kongelige 
Danske 
Videnskabernes 
Selskabs 
Naturvidenskabelige 
og 
Mathematiske 
Afhandlinger, 
vol. 
VIII, 
pp. 
61-144. 
Kjobenhavn. 


Fundelius, 
E.F. 
1972. 
Fossil 
vertebrates 
from 
the 
Fate 
Pleistocene 
Inglesidefauna, 
San 
Patricio 
County, 
Texas. 
Report 
of 
Investigations, 
Bureau 
of 
Economic 
Geology,TheUniversityofTexasatAustin77:1-74. 
Fydekker, 
R. 
1894. 
Contributions 
to 
a 
knowledge 
of 
the 
fossil 
vertebrates 
of 
Argentina 
111. 
An. 
Mus. 
Fa 
Plata, 
Sec. 
pal., 
11. 
118 
pp.

Marshall, 
F.G., 
A. 
Berta, 
R. 
Hoffstetter, 
R. 
Pascual, 
O.A. 
Reig. 
M. 
Bombin 
and 
A. 
Mones. 
1984. 
Mammals 
and 
Stratigraphy: 
geochronology 
of 
the 
continental 
mammal-bearing 
Quaternary 
of 
South 
America. 
Palaeovertebrata, 
Montpellier, 
Mem. 
Extra., 
1-76. 


Moreno, 
F.P. 
and 
A. 
Mercerat, 
1891. 
Exploracfon 
arqueologica 
de 
la 
Provincia 
de 
Catamarca. 
Paleon. 
Rev. 
Mus. 
Fa 
Plata 
1:222-234. 
Patterson, 
B. 
1967. 
Order 
Edentata 
in 
The 
Fossil 
Record, 
ch. 
30, 
Mammalia 
by 
P.M. 
Butler 
et 
al., 
Fondon 
(Geol. 
50c.):771-772.
Patterson, 
B. 
and 
R. 
Pascual. 
1972. 
The 
fossil 
mammal 
fauna 
of 
South 
America 
in 
Evolution, 
Mammals 
and 
Southern 
Continents. 
A. 
Keast, 


F.C. 
Erk, 
and 
B. 
Glass, 
eds., 
State 
University 
of 
New 
York 
Press, 
Albany,
247-309. 
Paula 
Couto, 
C. 
de. 
1954. 
Sobre 
um 
Gliptodonte 
do 
Urugai 
eum 
Tatu 
fossil 
do 
Brasil. 
Notas 
prelim, 
e 
estudos, 
Div. 
Geol. 
e 
Mineral., 
Dep. 
Nac. 
de 
Prod. 
Mineral. 
80:1-10. 


—. 
1956. 
Resumo 
de 
memorias 
de 
Fund 
sobre 
las 
cavernas 
de 
FagoaSanta 
e 
seu 
conteudo 
animal. 
Univ. 
do 
Brazil. 
Mus. 
Nac. 
Publ. 
Avulsas,
16:1-14. 



. 
1979. 
Tratado 
de 
Paleomastozoologia. 
Acad. 
Brasileira 
de 
Ciencias. 
Rio 
de 
Janeiro. 
590 
pp.

Porta, 
J. 
de. 
1962. 
Edentata 
del 
Mioceno 
de 
La 
Venta 
(Colombia). 
I. 
Dasypodoidea 
y 
Glyptodontoidea. 
801. 
Geol., 
Univ. 
Indus. 
Santader 
10:5-23. 
Robertson. 
J.S. 
1976. 
Latest 
Pliocene 
mammals 
from 
Haile 
XV 
A, 
Alachua 


County,Llorida. 
Bull. 
Llorida 
State 
Mus. 
Biol. 
Sci., 
20(3): 
11 
1-186. 
Romer, 
A.S. 
1966. 
Vertebrate 
Paleontology. 
3rd 
Ed. 
Univ. 
Chicago 
Press,
468 
pp.
Rovereto, 
C. 
1914. 
Los 
estratos 
araucanos 
y 
sus 
fosiles. 
An. 
Mus. 
Nac. 
Buenos 
Aires, 
25, 
247 
pp. 
Savage, 
D.E. 
and 
D.E. 
Russell, 
1983. 
Mammalian 
Paleofaunas 
of 
the 
World. 
Addison-Wesley 
Pub. 
Co. 
Inc., 
Reading, 
Mass., 
432 
pp.

Scillato 
Yane, 
G.J. 
1980. 
Catalog© 
de 
los 
Dasypodidae 
Fosiles 
(Mammalia,
Edentata) 
de 
la 
Republica 
Argentina. 
Actas 
II 
Congreso 
Argentine 
de 
Paleontologia 
y 
Biostratigraffa 
y 
I 
Congreso 
Latino-americano 
de 
Paleontologia, 
Buenos 
Aires. 
Vol. 
111:7-36. 


Sellards, 
E.J. 
1915. 
Chlamytherium 
septentrionalis, 
and 
edentate 
from 
the 
Pleistocene 
of 
Florida. 
Amer. 
J. 
Sci., 
4(40): 
19-31. 


—. 


1915. 
The 
pebble 
phosphates 
of 
Florida. 
Fla. 
State 
Geol. 
Surv. 
Seventh 
Ann. 
Rept., 
25-116. 
Simpson, 
G.G. 
1930. 
Holmesiua 
septentrionalisan 
extinct 
giant 
armadillo 
ofFlorida. 
Amer. 
Mus. 
Novit., 
442:1 
-10. 
, 


Webb, 
S.D. 
1974. 
Chronology 
of 
Pleistocene 
land 
mammals 
in 
Florida, 
pp.
5-31 
in 
Pleistocene 
Mammals 
of 
Florida, 
Webb, 
S.D., 
ed., 
Univ. 
Florida 
Press, 
Gainesville, 
270 
pp.