1

1

W!

'F
H

I“

U‘IIIWHI

OLA 1
01—3

 

24
I:

_cnooo1

v BEAN VERTEBRATES ”REM CINTQRL

70-2“ the Degree 0? M. S.
E‘JIECHESM STA”? *1?

I
.
huh.

ffifii
nut}.

EWE .SiTY

warm mam-i LEHE'Gm

‘1.

JIM“ WWWk]\fifi'fil‘xflfltfiufiflfimfifi\I * L/

3 1293 00671 7668

 

 

9:731; E U *9u;{\\i/,

ABSTRACT
ORDOVICIAN VERTEBRATES FROM ONTARIO
By
Kathleen Anne Lehtola

Vertebrate fossil fragments found in the northeast
part of St. Joseph Island. Ontario. Canada have been
referred to Astggspis desideggta walcott on the basis
of their stellate tubercles. The fossils are found in
shallow marine limestones of Middle Ordovician age.
This is the first recorded occurrence of an Ordovician
vertebrate in limestone. This find considerably widens
the geographic range of AstrasEis and lends some addi-
tional support to the theory of the marine origin of

the vertebrates.

ORDOVICIAN VERTEBRATES FROM ONTARIO
By
Kathleen Anne Lehtola

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE

Department of Geology

1971

ACKNOWLEDGMENTS

I wish to thank the following members of my MS com-
mittee for their assistance: Dr. J. Alan Holman (Chairman).
Dr. James H. Fisher (both of the Department of Geology.
Michigan State University). Dr. Jane E. Smith (Lyman Briggs
College, Michigan State University) and Dr. Gerald R. Smith
(Associate Curator of Lower Vertebrates. Museum of Paleon-
tology. University of Michigan).

Dr. Robert H. Denison (Chicago Field Museum) kindly
offered advice and encouragement during the course of this
study. I wish to thank Dr. Fisher for extra help in the
field and Dr. Robert Kesling (Museum of Paleontology. Uni-
versity of Michigan) for help in identifying fossil in-
vertebrates. Also thanks to Mr. Robert Kelley (Michigan
Geological Survey) who helped me locate some of.the older
reference materials, and to Mr. Karl Kutasi (Museum of
Paleontology, University of Michigan) and to Mr. Ralph
Taggart (Department of Geology. Michigan State Univer-
sity) who kindly photographed the fossils. Miss Nancy

Underwood made the maps in Figures 1 and 2.

11

TABLE

LIST OF TABLES. . . .
LIST OF FIGURES . . .
INTRODUCTION. . . . .
GEOLOGY . . . . . . .

SYSTEMATIC PALEONTOLOGI

DISCUSSION. . . . . .
LIST OF REFERENCES. 0

OF CONTENTS

iii

iv

11
20
27

LIST OF TABLES

Table Page

1 Stratigraphic relationships of
Trenton and Black River rocks. . . . . . h

iv

Figure

LIST OF FIGURES

Paleozoic rocks in the Upper Peninsula
of Michigan as mapped by Carl Rominger.
(From Michigan Geological Survey, 1873).

Eastern tip of the Upper Peninsula of
Michigan, including islands in the St.
Mary's River. The outcr0p area on St.
Joseph Island is solidly shaded. (Mo-
dified from Vanlier and Deutsch. 1958) .

Three views of an Astggspis plate.
Number 3:1 shows the plate in the mat-
rix. Plate is in the upper left cor-
ner. Numbers 3:2 and 3:3 are close-ups
of the plate in different aspects. 3:1
(I 4); 3:2, 3:3 (I 1#.2). Photograph
by Mr. Karl Kutasi. University of Mich-

lganooooooooooaaooaeeo

Five views of a group of tubercles from
the dermal armor of an Astraspis desi-
erata (UMMP V5797?) from St. Joseph
sland. Couplets 1-3 are stereo airs:
4-5 are single views. Numbers 1- have
been treated with silver nitrate. (x 20)
Photograph by Mr. Karl Kutasi. Univer-
lity Of MIGhigan.o-e o o o e e e o e e e

Astgaspis desidezgta; showing two views
of tubercles: :1 in cross section. and

5:2 a horizontally sectioned tubercle.
(x 165). Photograph by Mr. Ralph Tag-
gart. Michigan State University. . . . .

Paleogeographic map of North America.
Solid black areas are Middle Ordovician
outcrOps; cross-hatching is thought to
be the extent of the epicontinental
seas. (From Chamberlin and Salisbury.

1930). c o e o e e o o o e o e e e o e o

Page

. 18

INTRODUCTION

The earliest fossil vertebrates have perplexed several
generations of workers. The remains are fragmentary. not
at all abundant and, at times, poorly preserved. The ear-
liest recorded occurrence of Ordovician vertebrates was
made in Russia by J. V. Rohon (1889). who published a re-
port of Ordovician fish remains from Russia's Baltic re-
gion. He established their vertebrate nature on the basis
of the presence of pulp cavities. dentine and dentine tu-
bules, and enamel. These fossils have not been studied
further but are currently placed in the Order Thelodontia
by Russian paleontologists (flrvig, 1958).

Ordovician fish remains were next discovered in Colo-

rado and described by C. D. Walcott (1892). Re named two

 

genera and Species: ASLPBSBIS desiderata Walcott and g3;-
ptzchius americanus Walcott. The two genera are easily
distinguished, Eriptzchius having raised, elongate ridges.
and Astraspis having stellate or mushroom-shaped tubercles.
COpe (1893) suggested the fossils be placed in the Class
Agnatha. In 1936, M. L. Bryant did the first thorough
histological study of the two genera. He referred them
to the Order Heterostraci on the basis of (1) a prominent
median dorsal plate, (2) ornamentation superficially simil-
ar to that of other heterostracans and (3) aspidin

1

(acellular bone).

Darton (1906, 1909). Furnish et a1. (1936). and Ross
(1957) identified Ordovician ostracoderms in South Dakota,
Wyoming and Montana. Furnish and Darton both listed in-
determinate vertebrate remains. and Ross reported Eripty-
ggigg americanus and Astraspis desiderata from well cores
in Montana. Vertebrates of that area were first thorough-
ly studied by Orvig (1958) who named a new genus and species,
Pycnasgis Splendens ¢rvig, from near Sheridan, Wyoming.
Denison (1967) later relegated chnaspis to the synonymy
of AstrasEis, naming a second Species A. aplendens.

In 1958, G. w. Sinclair reported two Astrasgis plates
from British Columbia and Quebec in Canada. Unfortunately,
before these could be studied they were lost (Thomas E.
Bolton, Geological Survey of Canada, pers. comm.). Other
Cambrian and Ordovician vertebrates have been recorded at
various times (review in ¢rvig. 1958; Denison. 1967), but
all have been discredited or need considerable further
study.

A reported Ordovician ostracoderm occurrence that was
never discredited involved material from the Black River
Group near Escanaba, Michigan. In his Historical Geology.
R. C. Hussey (1947) figured a platelet of supposed verte-
brate origin. The figured specimen has been lost. I
micrOprobed and thin-sectioned another specimen (UMMP
Vh3936) identified as an ostracoderm by Hussey. It has
an outer covering of silica and a carbonate interior

(Robert H. Ehrlich, Michigan State University, pers. comm.).

3
But it has been so diagenetically altered that the organic
origin of the plate cannot be established.

In the course of my study of Hussey's Middle Ordovi—
cian collection from St. Joseph Island. Ontario, several
samples were found to contain vertebrate plates with tu-
bercles. These plates are here described and referred to
Astraspis desiderata.

Dr. Gerald R. Smith (Associate Curator of Lower
Vertebrates. Museum of Paleontology, University of Michi-
gan) granted me permission to study the fragments. They
are now again in the Museum of Paleontology collections
at the University of Michigan.

This is the third major recorded occurrence of Ordo-
vician vertebrates. It is the first occurrence of Astras-
pig outside of the Cordilleran region of the western United
States. It is the first occurrence of an Ordovician verte-
brate in limestone. The paleoecological significance of
the limestone is that it probably was deposited in the
Open sea, thus lending support to a salt-water rather than

a fresh-water origin of vertebrates.

GEOLOGY

Hall (in Hussey, 1936) first reported rocks of Middle
Ordovician age in the Upper Peninsula of Michigan. He cor-
related them with the Trenton Group in New York. Winchell
(1861) and Rominger (1873) traced these in outcrOps from
Wisconsin across the Upper Peninsula and into Canada (Fig-
ure 1).

The rock terminology has changed considerably in the
literature since Hall's report. Hussey (1936) broke the
Trenton Group down to the Black River and Trenton Forma-
tions. Martin (1956) elevated the Trenton and Black River

Formations to group status (Table 1).

 

TABLE 1. Stratigraphic relationships of
Trenton and Black River rocks.

 

 

 

 

 

 

Series Group Fogggtlog
Grogs Quarry
Trenton
Mohawkian Chandler Falls
Black River Bony Falls

 

The Trenton and Black River Groups are mainly lime-
stones and dolomites. The best outcrOps in the Upper Pen-
insula are around Escanaba. in Menominee. Delta. Marquette
and Alger Counties. At each of his 86 localities. Hussey
(1936, 1952) described the Trenton and Black River Groups.
subdividing the outcrOps into zones. Zones were differ-

entiated on the basis of lamination. pureness or foreign

u

FIGURE 1. Paleozoic rocks in the Upper Peninsula of
Michigan as mapped by Carl Rominger. (From Michigan Geo-
logical Survey, 1873).

 

_ maze—...

 

ZOE!
wxaj

 

  

 

<zuo m... ZEm
<ozozo

 

 

,V e

§z<
Io;—

wv<4

 

sow

 

 

b s
«ex
/ ‘9...”
.w . .

 

_ZO._.mm_2_n_ <m<o<_z
L r _ _ _ L a _ _

 

02442 «<0 m

mid

 

 

 

 

 

.mem pqaqm /

‘

><m

fljwtw .mflun.w...........m.m.......u ... ,. u.

 

’

f

LVI

 

    

moEmaam 92..

 

305‘:

.d

255.2«8? I

om?

 

 

 

 

 

 

 

er

 

 

 

 

 

7
content of the limestone. color. weathering. faunas and
contacts.

In differentiating the Trenton Group from the Black
River Group, Hussey (1936, p. 243) stated, “Fossils are
not as abundant in the Black River as in the Trenton and
are usually more difficult to obtain. The break between
the two formations cannot be detected upon lithologic
grounds but is entirely a faunal one. The relatively un-
fossiliferous Black River is succeeded by the more high-
ly fossiliferous Trenton with the typical brachiopod.
zzggspira gecurvirgstris at the very base."

The ostracoderm-bearing rocks of Black River age
are from St. Joseph Island, Ontario. which lies off the
eastern tip of the Upper Peninsula of Michigan (Figure 2).
The outcrOp area where Hussey collected fossils is on the
northeastern side of the island. Hussey (1952) stated,
"The tOp of the section begins at Quarry Point. on the
northeast side of the Island, and continues northward to
Gravel Point, and then westward for about two miles,
along the north shore of Landspur.“ The outcrOp area is
solidly shaded in Figure 2. Unfortunately. Hussey did
not pinpoint his collecting site.

Limestones from this outcrOp area are pure. fine-
grained. gray and very fossiliferous. A few brachiopods
have been completely filled by calcite crystals; some
invertebrate fossils have been partially replaced by

pyrite.

FIGURE 2. Eastern tip of the Upper Peninsula of
Michigan, including islands in the St. Mary's River.
The outcrOp area on St. Joseph Island is solidly shaded.
(Modified from Vanlier and Deutsch. 1958).

 

N manor...

 

 

m3...
w._._!o._oo n o
924

wZOthZD

07? w 252.35

92.4 -
£05537

%__H______
F _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VV. ._ai_a~_
Q

 

 

 

 

H
_
zflwsromwmm 1m are.

 

 

 

 

 

 

 

H 1 h H

 

   

 

 

 

q.___~___ ammsa . Q\U/

28% .zsoMSoomo 38.2

 

 

 

     

.4 enz<Emz<o e... zq
....... .. I”: .~.. mu.” . . w¥v$< ... . . .sfiWM-flnwvu ... H ”n... . .... . 50......mmmma

.0” mm. a a. t. .
Hesse... . . ...e...

... .......:.m. ................. .. . .. . . H a W a _
Dz<mawnowm ...».r. .u... “.... ..u .h a w a a n a H r . aLJ _ a

i

.)

 

o m 3.. ...H. .u ...u

 

 

10

The age of the St. Joseph Island rocks is put at Lower
Black River by Hussey (1936), making them slightly older
than any of the western vertebrate-bearing rocks. The Har-
ding sandstone of Colorado has been dated at Late Black
River or Early Trenton (Walcott. 1892; Kirk. 1930). Fur-
nish (1936) determined that the Black Hills were the same
age as the Harding. Darton (1906) called the Bighorns
Late Black River or Early Trenton. but Miller (1930) called
the Harding-equivalent beds Richmond in age. Ross (1957)
dated the vertebrate strata of the Williston Basin as
Early Trenton. The oldest known vertebrates still remain
those of Russia which are considered Lower Ordovician

(Denison, 1967).

SYSTEMATIC PALEONTOLOGY
Astras s deside ata walcott, 1892
Figures 3. h. 5)

Material.--Disarticulated plates with stellate tuber-
cles. UMMP V57977.

Age and localitz.--Middle Ordovician (Black River
Group). From fossiliferous marine limestone. "St. Joseph
Island: West end of North Channel in Lake Huron.” UMMP
Accession Card 1953 O-lh.

Diagnosis.--The fossils are assigned to the genus
Astraspis on the basis of tiny. discrete. disarticulated
plates with tubercles attached: the tubercles have a cent-
ral pulp cavity surrounded by laminated dentine. They are
assigned to the species 5. deslderata because the tuber-
cles are small (0.25-0.40 mm. in diameter) and stellate.

Remarks.--The ostracoderms occur in a nearly pure
limestone, with some detrital quartz. No very large frag-
ments are found, the largest having 16 tubercles in three
rows and measuring about 1 by 3 mm. (Figure 3). A few
single tubercles are found but most are in groups of three.
four or five.

The tubercles are soft and scratch very easily. A
weak zone may be present along the tubercle base horizontal
to the bony plate beneath because many tubercles have frac-
tures there that show air bubbles when dampened, and many
tubercles have broken completely off at that place.

The primary sample contains about one surface tubercle

per cm2.. and saw cuts in any direction through the plate
11

12

FIGURE 3. Three views of an Astraspis plate. Num-
ber 3:1 shows the plate in the matrix. Plate is in the
upper left corner. Numbers 3:2 and 3:3 are close-ups of
that plate in different aspects. 3:1 (x h): 3:2, 3:3 (x
14.2). Photograph by Mr. Karl Kutasi. University of
Michigan.

 

 

13

h:
a
p

 

m maze:

 

19

have usually revealed at least one tubercle.

Externally. each tubercle is stellate with a tiny
apical knob. and all have a definite pearly lustre (Figures
3 and #:5). No mushroom-shaped tubercles (cf. Orvig. 1958)
have been found among the specimens examined from Ontario.
The number of rays on each star varies between eight and
13. The maximum variation on any one plate is between
nine and 13. As tubercle size increases, the number of rays
increases (Sawin. 1959). The tubercles may have grown by
division of rays: Figure “:1 shows such a division on the
front tubercle.

The tubercles are irregularly but closely spaced on
the plates. Tubercle sizes are also variable. The size
range of tubercles on the plate in Figure 4 is 0.25 to 0.h0
mm. Tubercle size ranges for Wyoming specimens is 0.19 to
1.39 mm.. slightly less than Denison's (196?) maximum dia-
meter of 1.h5 mm. His minimum diameter for western Astras-
pl; is 0.10 mm.

The basal bony plate is extremely thin in the Ontario
specimens. The western vertebrates are. in general. much
thicker (Denison, 1967), but there is considerable varia-
tion in their thickness. Therefore. it is not known whether
the thinness of the Ontario plates is a reflection of the
area of the body from which they came. or whether the en-
tire animal had thinner armor.

The tubercles have central. round pulp cavities (Fig-
ure 5:1). These cavities are surrounded by dentine. Fine

laminations in the dentine parallel to the pulp cavity may

15

FIGURE h. Five views of a group of tubercles from
the dermal armor of an AstraSEis desiderata (UMMP V57977).
from St. Joseph Island. Couplets 1-3 are stereo pairs:
4-5 are single views. Numbers 1-u have been treated with
silver nitrate. (x 20). Photograph by Mr. Karl Kutasi,

University of Michigan.

 

FIGURE 1+

 

17

FIGURE 5. Astraspis desiderata; showing two views of
tubercles: 5:1 in cross section, and 5:2 a horizontally
sectioned tubercle. (x 165). Photograph by Mr. Ralph Tag-

gart. Michigan State University.

 

 

 

18

 

FIGURE 5

19

be seen on a few specimens (Figure 5:2). On one specimen
the laminations appear to extend to the outer edge of the
tubercle. but this cannot be determined absolutely. No
evidence of any amorphous "enamel-like” layer as reported
by Bryant (1936). Denison (1967), and others can be seen
on the outer surface of the tubercle.

The exterior shape of the tubercles from Ontario is
identical to that of western Astraspls desiderata. The
Ontario Astraspis have an extremely thin basal plate and
a smaller average tubercle diameter than those of the Cor-
dilleran region. Possibly the small average size indicates
that we have so far only sampled a few plates from young
individuals. The Ontario and western Astraspis may have
lived in different epicontinental seas or on different
coasts of a single sea. But until more information is
available about Astraspis and its distribution. nomencla-
torial recognition of the Black River form is not warran-

ted.

DISCUSSION

The fact that Ordovician vertebrates have been des-
cribed from three such widely separated localities as
Russia, Ontario and the western United States indicates
they are much more widely distributed in North America
than formerly thought and suggests the possibility of a
world-wide distribution. It is interesting to note the
occurrence of Astraspis desiderata from both the Cordil-
leran region of the United States and from Ontario, Can-
ada, marine localities about 1500 miles apart.

The question of a fresh or salt water origin of
vertebrates is an old but still controversial one. The
first paper to deal specifically with the subject was
that of T. C. Chamberlin (1900). who prOposed a fluvia-
tile origin for vertebrates based on the basic fusiform
shape. He believed the fish develOped their shape by
living in swiftly flowing currents. In other words, the
animal essentially was still and the medium moved past it.

In 1935. Homer and Grove published their classic
paper on the environment of the early vertebrates. They
discussed the paleontology of all vertebrate-bearing
strata in North America from the Ordovician through the
Devonian. Only their discussion of the Ordovician depo-
sits will be reviewed here because the Silurian and De-
vonian localities have no bearing on the origin of verte-
brates controversy.

The Ordovician Harding sandstone of Colorado is
20

21

considered to be a.marine deposit because it contains mar-
ine invertebrates. Because the vertebrate material. though
abundant, is always fragmentary and often water-worn. Romer
and Grove (1935. p. 810-811) concluded. ”The littoral na-
ture of the deposit combined with the fragmentary nature of
the fossil material has led to the general acceptance of
the theory that the Harding sandstone vertebrates were in
life inhabitants of fresh waters.” They postulated the
transporting of millions of fish fragments down rivers to
the ocean. The Russian occurrence of Ordovician vertebrates
is mentioned only briefly by Homer and Grove. These fos-
sils occur in the marine glauconitic sands of the Baltic.

In rebuttal to Chamberlin's arguments put forth favor-
ing a fluviatile vertebrate origin. Berry (1925) stated
that when fish move through quiet water the same stresses
are put on the body as when the water moves past the fish.
He also said that, while most existing fish are fusiform.
that simply happens to be the shape which slips through the
water with the least resistance. Thus fusiform shape cannot
be considered as proof of fresh-water vertebrate origin.

Homer and Grove's conclusion as to the fresh-water
origin of the Harding vertebrates was based primarily on
the fragmentary nature of the remains. Denison (1956)
questioned their reasoning on several grounds: (1) some
of the fragments are water-worn, but many others are very
well preserved, (2) the fragments are found in equal
abundance at all outcrOps of the Harding, covering some

15,000 square miles, and at the Harding-equivalent outcrOps

22

in the Wyoming, South Dakota and Montana area, and (3)
the fragments are often quite large. up to about 4 cm..
yet the sand is very well sorted. Currents that can sort
sand grains so well could have sorted the vertebrates

had they been transported. Both Denison and Robertson
(1957) disregarded the bulk of Romer and Grove's paper
which was a detailed summary of the North American De-
vonian deposits. By Late Silurian or Early Devonian.
evidence is found for the existence of marine and fresh-
water vertebrates; thus. by then they are adapted to both
modes of life.

Nils Spjeldnaes (1967) would attribute the fragmenta-
tion of the vertebrates to bacteria and other scavangers
living in the sediments. Another possible explanation
for the fragmentation might lie in their suturing. The
anterior dermal armor of Astraspis and griptychius was
composed of tiny. discrete plates. The plates abutted
against one another with vertical sutures (Denison. 1967).
Perhaps in life the armor was held together by some or-
ganic cementing agent. When decay set in after death,
the plates would become loose and subject to any water
movement. This might account for the fragmentary nature
of the Ordovician vertebrates.

The new Ontario limestone locality is important as
the first vertebrate occurrence in limestone from the
Ordovician. Up to now all Ordovician vertebrates have
been reported from sandstones. shales and siltstones.

The Harding Formation in Colorado is a fairly fine-grained,

23
well-cemented, red sandstone and the Harding-equivalent in
the Wyoming area is a coarser grained, poorly cemented sand.
Darton (1909) reported some fish from the Black Hills of
South Dakota occurring in the Whitewood limestone. Fur-
nish et. al.(1936) figured the Black Hills vertebrate frag-
ments, but placed them in the underlying Harding-equiva-
lent Roughlock siltstone. Ross'(1957) reported fragments
in the sandstones and shales of the Winnipeg Formation in
the Williston Basin. The Russian vertebrates are in mar-
ine glauconitic sands.

Sandstones, shales and siltstones can occur in marine.
brackish or fresh-water. Limestones are nearly always
marine in origin; they imply a quiet environment. either
shallow or deep (Pettijohn. 1957).

The Ontario ostracoderm-bearing limestone is consid-

ered to be a fairly shallow cpen sea deposit (Hussey,
1936; Dr. R. Keeling, University of Michigan. pers. comm.:
Dr. C. E. Prouty, Michigan State University, pers. comm.).
Paleogeographic maps (Figure 6) indicate that Michigan was
under an epicontinental sea in the Middle Ordovician.

Invertebrates identified from the limestone which bears
the ostracoderms, include Liospira cf. eugenia (a low-
spired gastrOpod of the Ordovician), gzgggplgg‘;ggugzlggg-
ELLE (a brachiopod ranging from Middle to Upper Ordovician).
hagchotrema mlnnesotensis (a brachiOpod ranging from Mid-

dle to Upper Ordovician), hanchotrema sp.. ggleperditia
fabglites (an ostracod from the Middle Ordovician). a rib-

bon-like cryptostome bryozoan, and a colonial coral. No

24

FIGURE 6. Paleogeographic map of North America.
Solid black areas are Middle Ordovician outcrops: cross-
hatching is thought to be the extent of the epicontinen-
tal seas. (From Chamberlin and Salisbury. 1930).

J‘s—Fa—

 

 

25

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

no :60 150 I40 Isolzouomeoeo 10 so m 40 so 70
\ 6 -
.- 4‘9 Km”?
0 g l/
/
”afi. / / “ .
'Q‘ a .
i I
43‘ \
8’ .I
! 7 '
~_ ‘ I
. .4 I__\
x
.4-
’I
. o . . /’
d’b ,/
\‘\‘ < 3“. ’/
‘\ 5“ 0 . J ’ ” .tt'.“ 0
‘~~ _ 5 I; --_~ I.— 6.. .a‘:: f...
) . ". " ‘0
’ ° ' cc: .
\ . o :
° \(3 ' °.' ‘ ’
' /
”h. 51
no up no 9L r w 'm/{§h

 

 

 

FIGURE 6

 

 

2.6
trilobites are present. The fossils are well preserved
generally; some are fragmented, many are not. ggzgcho-
tram; is most abundant: Liosgira. Eoleperditia and the
bryozoan are also common.

In the west, Astraspis has always been found in asso~
ciation with Eriptychius. Eriptychigs has not yet been
found in the Ontario rocks.

Robertson (1957. p. 158) says that it is universally
agreed that corals and non-lingulid brachiOpods have
always been marine. He adds that bryozoa and gastrOpods
are predominately marine, and no known fresh-water record
exists for fossil bryozoa. Therefore. the limestone and
its invertebrate fauna would suggest a normal marine en-
vironment, quiet. fairly shallow. with a muddy or calcar-
eous bottom (Dr. H. Scott, Michigan State University,
pers. comm.: Dr. R. Kesling, University of Michigan. pers.
comm.).

The appearance of vertebrates in a normal marine,
cpen sea environment suggests strongly that they lived

there and were not introduced after death.

LIST OF REFERENCES

Berry. E. W. 1925. The environment of the early verte-
brates. Amer. Nat. 59:354-362.

Bryant, W. L. 1936. A study of the oldest known verte-
brates. Astras is and Eriptzchius. Proc. Amer. Phil.
SOC. 76( 8 09' 2?-

Chamberlin. T. C. 1900. On the habitat of the early verte-
brates. Jour. Geol. 8:#00-u12.

Chamberlin. T. C. and R. D. Salisbury. 1930. College text-

book of geology. Pt. II historical. Henry Holt and
Co. New York.

COpe E. D. 1893. The vertebrate fauna of the Ordovician
of Colorado. Amer. Nat. 27:268-269.

Darton. N. H. 1906. Fish remains in Ordovician rocks in
Bighorn Mountains. Wyoming. with a resume of Ordovi-
cggn geology of the Northwest. G.S.A. Bull. 1?:5h1-
5 .

. 1909. Discovery of fish remains in Ordovician of
tgg Black Hills. South Dakota. G.S.A. Bull. 19:56?-
5 .

Denison. R. H. 1956. A review of the habitat of the ear-
liest vertebrates. Fieldiana: Geol. 11(8):361-457.

. 1967. Ordovician vertebrates from western United
States. Fieldiana: Geol. 16(6):131-192.

Furnish. W. M.. E. J. Barragy and A. K. Miller. 1936.
Ordovician fossils from upper part of type section
of Deadwood formation. South Dakota. A.A.P.G. Bull.
203 1 329-131‘1 o

Hussey. R. C. 1936. The Trenton and Black River rocks of
Michigan. Mich. Geol. Surv. Pub. #0. ser. 3“. pt. 3.

. 19h7. Historical Geology. McGraw-Hill Book Co.
New York.

. 1952. The Middle and Upper Ordovician rocks of
Michigan. Mich. Geol. Surv. Pub. #6. ser. 39.

27

28

Kirk. Edwin. 1930. The Harding sandstone of Colorado. Am.
Jour. Sci.. 5th ser.. 20:956-465.

Martin. H. M. 1956. An index of the geology of Michigan:
1823-1955. Mich. Geol. Surv. Pub. 50.

Michigan Geological Survey. 1873. Atlas accompanying re-
ports on the Upper Peninsula. Julius Bien. New York.

Miller. A. K. 1930. The age and correlation of the Big-
horn formation of northwestern United States. Am.
Jour. Sci.. 5th ser.. 20:195-213.

Orvig. Tor. 1958. Pycnaspis splendens. new genus. new
species. a new ostracoderm from the Upper Ordovician
of North America. Proc. U.S.N.M. 108:1-23.

Pettijohn. F. J. 1957. Sedimentary rocks. Harper and Row.
New York.

Robertson. J. D. 1957. The habitat of the early vertebrates.
Camb. Phil. Soc. Biol. Rev. 32:156-187.

Rohon. J. V. 1889. Uber unter-silurische Fische. Bull. Acad.
Imp. Sci.. St. Petersburg. new ser. 1(33). no. 2:
269-277.

Romer. A. S. and B. H. Grove. 1935. Environment of the early
vertebrates. Amer. Midl. Nat. 16:805-856.

Rominger. Carl. 1873. Paleozoic rocks of the Upper Penin-
8‘1130 M1011. 08010 Sun. v0 1. pt. 30

Ross. R. J. 1957. Ordovician fossils from wells in the
:illiston Basin. eastern Montana. U.S.G.S. Bull. 1021-M:
39-510.

Sawin. H. J. 1959. Notes on the origin of bony armor. Rice
Inst. Pamphlet. 46:90-108.

Sinclair. G. W. 1958. Occurrence of fish in the Ordovician
of Canada. G.S.A.Bull. 69:1644.

Spjeldnaes. Nils. 1967. The paleoecology of the Ordovician
vertebrates of the Harding formation (Colorado. USA).
Colloqaes Tat. Gent. natn. Rech. Scient. 163:11-20.

Vanlier. K. E. and Morris Deutsch. 1958. Reconnaissance of
the ground water resources of Chippewa County. Michi-
S‘ne HIOh. 60010 Sun. Progress Rpt. 17.

29

Walcott. C. D. 1892. Preliminary notes on the discovery of
a vertebrate fauna in Silurian (Ordovician) strata.
G.S.A. 3111].. 33153-172.

Winchell. Alexander. 1861. First biennial report of the
progress of the geological survey of Michigan: 1859-
1860. Mich. Geol. Surv.

MICHIGAN STRTE UNIV LIBRQRIES

IIIIlIII1IIII2IIII|III|IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

 

293006717668