0
mil
2?.
”a
b j I
U

THE SEC: MERCANCE Oi:
CORN. SE‘TECM:

T133332 {m ‘m. Saga-a 5% M..

is S
MASEEGAQ 3TATE 363;...3-42

2ft! 3: -,~-; :v-r':E ’3'" .fa- --
align La" 1.23“; aigt‘x‘é‘i

THE-'3‘!

 

s. I
_-.,L- e- 5. -A h.-.‘ -- -‘ a - On 0. J an- .- an. .a‘- o a .——.—— -—--—-- - *-—l—_.—

This is to certify that the

thesis entitled

”The Significance of Oriented
coral Sections"

presented by

Ian-en mm Grabau

has been accepted towards fulfillment
of the requirements for

[actor's degree“! Geology

Major professor

pm In: 25. 1950

 

0-169

 

THE SIGNIFICANCE OF ORIENTED CORAL SECTIONS

by warren Edward Grabau

A THESIS

Submitted to the Graduate School of Mich igan
State College of Agriculture and Applied
Science in partial fulfillment of the
requirements of the degree of

MASTER OF SCIENCE
Department of Geology
and Geography
1950

THESIS

0leT ENTS

Acknowledgments . . . . . .
Introduction and Purpose . . .
Stratigraphy . . . . . . .
Techniques of Preparation . . .
Definitions of Orientation .
Interpretations of Structure . .
Carinae . . . .
Walls . . . .
Septa . . . .
Septal Dilation .
Size and numbers of

Dissepiments . .
Orientation Criteria.

emcee.

epta

Area Nomogram . . . . . . .
Conclusions . . . . . . .
Bibliography . . . . . . .

Appendix (Derivation of Nomogram)

(ii)

0 O 0 O O O O

Page

Q0303!“

PLATES

LO 08131 on I‘n’Iap o e e e e o o e e 0

Index to Location Map . . . . . . .

Plate
Plate

Plate

Plate
Plate
Plate
Plate
Plate

Plate
Plate
Plate

Plate

Plate

Plate

I: Types of Carinae and Implantation

II:

III:

Carinae MOdifications Due to
Orientation
Carinae masked by Dissepiments

Section of wall Between Two
Corallites

Relation of Septa to Well
Serration

Pseudo-serrate wall . . . .
Growth Rings and Septal Dilation
Areas and Numbers of Septa. .
Areas and Numbers of Septa. .
Dissepiments . . . . . .

Pseudo-concentration of Dissepi-
me nt S Q O O O O O O O

Dissepiments . . . . . .
Nomogram. . . . . . . .

Orientation of Sections
Types of Carinae . . . . .

Types of Carinae . . . . .
Extended Carinae . . . . .

Implanted Carinae ,

Carinae in Transverse-oblique
Section

masked Carinae . . . . .

Carinae in Longitudinal-oblique

section
walls
Growth Rings . . . .‘ . .

(111)

20

24

27
27
30

32
34
59

46
47

48

49

PLATES (cont.)

Page
Plate XV: Area-septa Correlation
Dissepiments and Tabellae
Compound and Simple Dissepi-
ments . . . . . . . . 50

Plate XVI: Dissepimental Chambers
Dissepimental Size Range . . 51

Plate XVII: Dissepiments in transverse-
oblique Section . . . . . 52

Plate XVIII: Serial Section . . . . . 53

(1V)

ACKNOWLEDGMENTS

I am indebted to Dr. William.A Kelly, of
Midiigan State College, for his considerable
assistance and cooperation. To Arthur Slaughter
I am obligated, because of his enthusiastic
collection of samples from areas which I was not
able to reach. I am.grateful to Dr. B. M.
Stewart, of the mathematics Department, for
his aid in the solution of the homogram included
in this paper. M1. E. J. wade is largely respons-
ible for the photographic material included.

(V)

INTRUDUCTION AND PURPOSE

In the examination of fossil animals, paleontologists
are forced to rely on only a portion of the evidence
available to biologists who study modern faunas. Specific
determinations of contemporary animals are made on the
basis of both soft and hard parts. Because paleontologists
are restricted to skeletal characteristics alone, it is
inevitable that their determinations will not have the
precise values possessed by identifications made from
soft parts as well as skeletons. This is especially true
in those cases where skeletons exhibit convergent, or
parallel features, presenting similiar characteristics
in quite different groups. Closely allied species may
have nearly or entirelyidentical skeletons, and paleon-
.tologists cannot be expected to identify them as indi-
viduals.

Mbdern biologists, who work with the living fauna of
the planet, are accustomed to the notion that wide vari-
ation in morphological characteristics is possible within
a given species. For instance, fishes of the same species
not infrequently have different numbers of vertebral seg-
ments. Corals exhibit quite different corallum shapes.

within the same species, and.within very narrow areal limits.

(1)

These differences, in most cases, can be correlated with
environmental factors (vaughan, 1917).

The question was, how does one detect the difference
between a variation within a species and a variation due
to different species? The most obvious approach was to
collect as many specimens as possible, and examine them
for overlapping characteristics. In theory, there should
be a consistent difference of some nature between two
species. But if the forms are only variations with species,
there ought to be a fairly consistent overlap of character-
istics, such that Specimens would grade imperceptibly into
each other. A plot of variation limits should then'be
possible, and any breaks in the sequence could then be
interpreted as specific breaks (Simpson, 1987).

As a complicating factor, the theory suggested that
some variation would be found within.a single corallum,
because the environment would not be the same over the
entire colony. For example, food supplies might not be
uniform.because one side of the colony would be in.a
,more advantageous position.mith respect to the food-bearing
currents. In the domal colonies, some corallites at the
periphery would be bothered more by mud and absence of
light than would those near the apex. These considerations
made it necessary to examine each corallite of every colony

individually.

To insure that no portion of the coral skeleton was
neglected, a system of oriented sections was designed.
A series of four sectionS‘was found to be most useful.
Study with the oriented sections revealed that the
structure of the coral animal is exceedingly complex.
Gradually this line of endeavor became a technique for

the study of Hexagonaria, and it is this technique which

 

furnishes the body of this paper.

The descriptive terminology used in this paper is
that used in Sanford's (1939) excellent summary of the
families of tetracorals. Where new or modified terms
are used, they are defined in the text, or illustrated,l

OI‘ bOthm
STRATIGRAPHY

The fossil corals used in this study were collected
from.outcrops of the Traverse group in the north-central
part of the lower peninsula of Michigan. The locations
of the various outcrops are noted on the accompanying map
of portions of Cheboygan and Presque Isle counties. meat
of the specimens were collected during the summer of 1949
by the writer, and by members of the student body of Mich-
igan State College while on a field trip to the area.
Additional material belonging to the collection of Henry
Faul was later included.as better techniques for the inter-

pretation of random.sections were developed.

(3)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

one _ mNm m m_m _ szm _ awe
I a.
> , Egon Steam
1 ‘1; \ INVH NV h
C. . II.
.N. 138334.: 338 see r»; 1 _ _|1
~ F X __ WV J m . .2330 £322wa
1,]. I L/ >41.ch e8 FEM _ J/ zoEmJ
_ _ .
Mr v}— _ . I \ r 323% 396.0.”MW .3330
. , , COskq
flBu
aw» 33m ASE xzok m_> .Nm. VNWMW .Cuoao PBSSMV .— _
led III! oz<m0un
l 08:80 an _ m Emmet 8%de
Q . .
m 6828.0 If... a. _
p . hem Slot U \ W
. 5:030 33 «85 .i ,
quq x0615
mud: z. mn<om
— — 1 d. d W . in.—
o. m . o

l
s .
m. 6202.8 one; mcoEBeam 52;; So: 20:83 NV

 

 

 

 

 

<mm< m.._.._<n_ OOmDOOO I 20.7.4

 

 

no.2 xwoz.

I NbEl

N952).

 

 

INDEX 29 LOCATION MAP

Location Specimen Location Formation
number number
Afton Quarry 200 5 Gravel Point
' ” 100 5 Gravel Point
' ' 13-1** 5 Gravel Point
” ” 14-1 5 Gravel Point
Legrand Query 400 9 Gravel Point
" ' 450 9 Gravel Point
Draper School 21-1 28 Gravel point
Bunker Farm 7-1 14 Gravel Point
Gorbut School 11-1 9 Gravel Point
Beebe School 2-2 1 Gravel Point
” ' 2-1 1 Gravel Point
Afton Quarry 25 5 Koehler
' ' 30 5 Koehler
" ' 35 5 Kbehler
Silva's Folly 15 12 Genshaw
Tower Dam. 18-1 21 Genshaw
' ' 18-2 21 Genshaw
' ” 18-4 21 Genshaw
Rainy River Falls 29-1 13* Genshaw
unknown location 24-1 Genshaw
unknown location 23-2 Genshaw
7 15-1 7 Genshaw
Black Lake Quarry 500 34 ‘Ferron Point
" ' " 25-8 34 Ferron Point
' ' ' 25-5 54 Ferron Point
Ocqueoc Falls 75 51 Ferron Point
Rockport Quarry 500 Rockport
' ' 550 Rockport
Kalkaska County*** 14558 Traverse

Andditional eight specimens were used, but it is
known only that they came from.the Afton-Onaway area.
The location numbers used in this paper'are Uhose estab-
lished by Kell and Smith (1947).

*Location numbers marked wifii an.asterisk are those
established by verhocven (1948).

**The hyphenated numbers represent specimens collected by
Henry Paul (1942).

***This specimen is from the cuttings of an oil well drilled

in Kalkaska County. The number given is the SP’number of
the well, and is on file with the Michigan Geological Survey.

(5)

The collection includes corals from the Rockport
Quarry limestone, the Ferron Point Formation, the Lower
Genshaw Formation, the Killians member of the Genshaw,
the Koehler limestone, and the Gravel Point formation,
listed in ascending order. Some 35 different specimens
were included in the study. However, certain forms were
studied by oriented serial sections, so that the total
number of sections examined is approximately one hundred.
The 35 specimens are not distributed uniformly through
the stratigraphic column, but instead tend to be concen-
trated in those formations or members which are most
fossiliferous, or in which the most diligent search was
conducted. Thus, the Gravel Point is a point of concen-
tration because itb outcrops were thorougi 1y searched,
and because it is fossiliferous. In contrast, only a few
specimens from the Koehler are included because of the
lack of fossils in that formation. Some of the Specimens
from the Lower Genshaw are so badly crushed as to be

almost useless.
TECHNIQUES 9}: PREPARATION

The most common method of studying corals is by the
use of thin sections. However, the writer finds that
sections of standard thickness (0.02mn) are not the most

useful, because they are too thin to reveal third dimension.

(6)

The thin sections used in.this study are from 0.06mm to
O.lmm.in thickness, depending on.the method of preserva-
tion. In those specimens preserved in clear calcite, the
sections were made thick, so that the maximum.possible
relief could be obtained. These are the most satisfactory
for direct study, but it is difficult to obtain good
photographs of them, If the fossil is preserved in clay
minerals, or in calcite sufficiently impure as to be
opaque, the slides were cut thinner, but only enough to
allow the passage of light. The retention of enough
thickness to allow three-dimensional interpretation is
of first-order importance. This is especially true in
those specimens where precise control of orientation is
difficult or impossible.

It was not uncommon to have corallums break into
fragments when an attempt was made to cut them.on a
diamond saw. It was found that there are systems of
internal fractures, along Which the fossil will part at
the slightest strain. To avoid this, the specimens were
soaked in liquid resin for from.one to two hours.
Usually the resin permeated the fractures,and the coral,
when cool, was firm.enough for handling and processing.

Four different orientations of sections were used.
The transverse section cuts the axis of the corallite at
approximately 90 degrees. It is difficult to obtain a

completely accurate transverse section, and so the term

is defined as being a section which cuts the corallite
axis at an angle between 85 degrees and 90 degrees.

Transverse-oblique sections are defined as being
sections waich cut the corallite axis at an angle between
70 degrees and 85 degrees.

The lpngitudinal section is defined as a plane which
either contains the axis of the corallite, or is parallel
to the axis. When the section contains the corallite
axis, one much-to-be-desired result is the display of
septa lying in the plane of the section. Ordinarily the
septa are sufficiently sinuous so’that at least a portion
of one septum is brought into the plane of the section,
even though other portions of the section may pass between
two septa. Most corallites are somewhat sinuous, and so
it is virtually impossible to obtain an ideal longitudinal
section for distances along the axis of more than.a few
millimeters. The usual case is to find perhaps a centi-
meter of longitudinal section with the section becoming
more or less oblique on either side as the axis of the
corallite bends out of the plane of the section. Almost
all longitudinal sections are in planes parallel to the
axis, instead of coinciding with.the axis. Such sections
are valuable, for they are useful in the interpretation

of carinae and dissepiments.

(8)

The longitudinal-oblique section is defined as a
plane which intercepts the axis of tie corallite at an
angle ranging from five degrees to 30 degrees. The
sinuosity of the corallites usually results in a short
area of longitudinal section, with longitudinal-oblique
portions of the sane corallite both above and below the
longitudinal portion.

Most of the Hexagonaria from.the Traverse group are
forms with a dome-shaped, or hemisheroidal corallum.

The diameter of the colonies averages perhaps 20 centi-
meters. Some are very much larger, as for example the
specimen found at the locality called Silva's Folly. The
entire corallum.is unknown, but it must have been fairly
large, for no perceptible radiation of the corallites from
a common center was evident, and yet the fragment obtained
had a volume of perhaps 3,000 cubic centimeters.

For these dome-shaped corallums, a standard tech-
nique was developed. The sample was first cut in half
so that the result was two half-hemispheres (See Plate XI).
Usually such a cut revealed a series of mere-or less
longitudinal sections. If the cut was at an angle to the
desired orientation, the face of one of the hemispheres
was ground until it was a true longitudinal section. This
hemisphere was then placed in.the diamond saw and a thin

slice was cut off, mounted, and labeled as to orientation.

(9)

The other half-hemisphere was then ground to bbtain a
longitudinal-oblique section oriented at a very shallow
angle. This was then sawed as before, and the slice
mounted.

Because of the growth habits of these forms, it is
impossible to get more than a few corallites in transverse
section on one slide, because the radiating habit takes
the plane of the section ut of the normal very rapidly.
However, large sections have distinct advantages, in that
they allow both transverse and transverse-oblique sections
of the same corallum.to be examined.at the same time.
Therefore, one of the half-hemispheres is then placed in
the diamond saw such that the plane of the saw-cut will
be a chord of the are representing the surface of the
original corallum. Sereral thin Slices are then removed
and mounted as before. Only the central portion of such
sections will be transverse; the entire periphery becomes
more and more oblique. The remainder of the corallum.was
usually ground and polished, and given no further processing.

If the entire corallum was not available, the fragment
was examined carefully to insure a knowledge of the
direction of the coral axes, and a cut was made normal to
this direction. If the proper orientation was not achieved

at the first attempt, the surface was ground on a lap

(10)

(II)

°s4uemtdossrp kq pemro; sour: and with tectquepr sea qotqm
steqsfxc one neouqeq curt a pouoqs teed DGQSIUIJ oqq ‘itnaox
a sv °stsqs£10 sun JO seesaw upon one ueqq ftptdex
910m pestossrp sereteo JO stsqsfirc qusoefpe one ueenqoq
qoequoo and cone pone; sum 1; ‘9110190 or ponieserd
suomroeds uI :qsrxo con prp qotqn sernqonxqs'EEEEB
SGWTQGWOB steed neon Irenoostp and sea Sutureouoosrp fizeA
°asot ion ueqq uoqxo exom oxen ‘esurrso JO squomtdesstp
so none ‘soanoniqs Item: IxeA °(Lp5t ‘uuens) 'Aroqoe;
-stnesun fitsqetdmoo qsomte oq_oq.puno; 919M steeg
'squemSBJJ poduetroun on Ituo pet ooeians eietrdoxdde sun
Sutpuufi 48H; inomtm suomsqueno pensep em Sup sumo is
sidmoqav 'eoejrns qsqq oi tattered JJO sexed carve a pus
‘uotqequerro pexrnbez and data punoxs cert; SBM.GOBJInS
e ‘pozrsop erou.suorqequetio Joqqo ueqm °ooeyxns punch
em on tenured 5:0 named a]: com a pus ‘sreedde notices
teurpnqrfiuot a ttqun punch s; seosrzns eqq JO euo ‘os
JI °onbrtqo qeqmemos eq TIIM notices use one done pane;
Attensn a; a; ‘stxs eitttexoc sun on temxou firestoexd eq
qou fem notices esronsueri sue esneooq °uotioes Ieurpni
-;8uot e at GOBJIUS esxsnsuexq sun on tomrou oustd nus
esneoeq ‘suorqoes teurpnqreuot one no; eprnB e se scares
non eoexzns ssxonsusrq sum 'peaunom pus penomox sorts
eqq pus ‘eoexxns punoxs one on tattered open sea due a

stun Surmottog 'peureqqo ssmtuorqoos osxonsuerq a trash

With this discovery, the whole technique was hastily
abandoned.

Polished sections are in most cases the most re-V
vealing of all methods of study. When specimens pre-
served in color-microscOpe, the impression is one of a
casting in clear glass. Inclined light enables the
student to see deeply into the structure, and thus
relationships are revealed which are almost entirely
masked in thin sections. It was with the study of
polished sections that many of the intricacies of structure
were first detected. These discoveries were then used to
establish a set of criteria for use with thin sections,
and of such nature that the approximate orientation of
randmm sections could be determined. Before these cri-
teria were developed, only those sections for which exact
date on orientation was available were useful. But with
the orientation criteria, it was possible to expand the
base of investigation considerably, for they made it
possible to utilize already-existing thin sections for

which no data on.exact orientation was available.
mTERPRETArI CNS 93 STRUCTURE

Carinaez-Amond the structures of Hexagonaria used
for the identification of species are the carinae. These

are very small, linear masses of skeletal tissue which

2)

grow on the septa. (Twenhofel & Shrock, 1935) Carinae
are universally present in Hexagpnarig, No specimen,
upon careful examination, has been found to lack carinae
in any part.

In sections cut normal to the slope of the carinae,
they are seen to be an isomorphous series. Of the various
forms, the rarest is the crenulate carina, which is a
sharp-crested fold in the septum. The form is established
without appreciable thickening of die septum (See Plate I-a
and Il-b).

When a septum makes a low-angle turn, there is not
infrequently developed a thickening of the septa, so that
the flexure is reinforced by additional skeletal material.
This is the glbgw_type. (See Plates I-a and.Xl-b)

The modified 2132! form is a half-lozenge grown on
one side of a septum only, and with the long dimension of
fine lozenge parallel to the septum. (See Plates I-a and
XI-b)

The lozenge carina is a diamond-shaped structure with
its long axis parallel to the septum, and with the points
projecting on both sides of the septum. (See Plates I-a and
11-h)

Still another form looks like the cross-beam of a
telegraph pole, or the yard-arm on a ship's mast. (See

Plates I-a and.II-b)

(13)

In longitul inal section, the carinae exhibit similiar
diversities. The forms range from straight-line linear
occurences, through sigmoid, arched, and even bow-shaped.
(See Plates I-b and XII) The form which the carinae will
display in transverse section cannot be determined from a
lognitudinal section. As a corollary, the form displayed
in longitudinal section cannot be determined from transverse
sections. Any combination of transverse and longitudinal
forms is possible.

The angle at which the carinae intercepts the wall
of the corallite is called the m_u_r_a_l_.. 93312, while the
angle at which the carinae extended joins the axis of
the corallite is called the gig}; gm. (See Plate I-b)
The mural angle varies from 90 degrees to about 20 degrees,
and the axial angle varies from 90 degrees to approximately
40 degrees.

Several types of carinae can be found on a single
septum. Thus, for example, yard-arms may grade into either .
elbow types, or offset yard-arms, or both. (See Platen-b)

Although carinae seem to be strengthening structures
on the septa, they are not restricted to the septa, but in
some cases project well beyond the axial termination of
the septa. Such occurences are best seen in longitudinal-
oblique sections, where the extended carinae appear as

dots in a row along a line formed by the extension of one

.4)

of the septa above or below the occurence. (See Plate XII-b)

In longitudinal sections there are many examples of
the implantation of a secondary carina between two pre-
viously existing, or primary carinae. (See Plate 1-0 and
IXIII-a) The primary carinae reach from the corallite wall
to the end of the septa, or in some cases beyond, but the
secondary carinae are.ordinarily present only in the vicinity
of the tabularium, and disappear well before they reach
the wall. Counts of primary carinae along a wall will
result in a lower count than one taken along the edge of
the tabularium, where it will include both primary and
secondary carinae. (See Plate I-c)

Carinae counts can yield varied results if the orien-
tation of the section is not considered. As an example,
(See Plate I-c) a transverse section may result in a low
count, because it cuts across the carinae series at an
angle. Furthermore, in transverse section it is possible
that no carinae will show near the tabularium, because in
those highly arched forms in which the axial angle
approaches 90 degrees the carinae will be parallel, or
nearly so, to the plane of the section. But carinae will
show more or less abundantly near the periphery, leading
to the erroneous conclusion that carinae exist only near
the walls, and that the axial ends of the septa are free

of carinae. If the angle at which a transverse-oblique

section is cut happens to be the same as the slope of the
carinae on one side of the corallite, that side will be
without apparent carinae, while the other side of the
carinae will be abundantly carinate, because they will
have been out almost normal to their slope. (See Plates
1-0 and XIII-b) A corallite which presents abundant
carinae on one side and none on the other cannot be
interpreted on that evidence alone as being non-carinate
on one side. (Stumm, 1949)

Carinae which are cut at shallow angles can easily
be mistaken for minor dilations of the septa, and omitted
from carinae counts for that reason. This is especially
true for the elbow, lozenge, or crenulate types, inasmuch
as they are short and stout to begin with, and the dis-
tortion produced by the angle of interception of the
section increases the length, but not the width. (See
Plate II-a)

In some places carinae are partially masked by
dissepiments, and consequently overlooked in transverse
or transverse-oblique sections. There seems to be no
inter-relation.between carinae and dissepiments. If a
section happens to pass through a point where a dissepi-
ment and a carina cross, the appearance will be only that
of a slight thickening of the dissepiment where it joins
the septum. (See Plates II-a and XIII-c)

t

(16)

 

 

 

 

PLATE I
TYPES OF CARINAE
TRANSVERSE SECTION LONGITUDINAL SECTION
/——-———Crenulale +—-Arched
:«Slgmold
I

«+— Bow - shaped

  

4—Modified elbow
/ I

‘ I
l
I
I
I
Linear I/
, {N
' .
, anal angle
I I
I
I
, I
I
I

 
 
  

4— Lozenge

 

«Yard—arm mural on Is I

+Offsel yard-arm axis

seplum septum

wall

 

 

 

IMPLANTATION OF CARINAE and CARINAE COUNTS.
LONGITUDINAL SECTION

wall axis im lanled carina
/ P

.4/
/ ,Transverse-obllque section:
//. carinae counl = |,'

.. , Transverse section:
carinae count = 3.

/'
I-Longiludinol section:

é carinae count = 8 per inch.
I’I- Longitudinal section:

carinae count = 5 per inch.

 

’4

 

 

 

 

 

 

 

 

 

 

 

 

 

l7

 

Some uniform method of counting carinae would be
useful. It is suggested that such counts be made from
longitudinal sections only, and only in directions normal
to the slope of the carinae. Even this must be used with
discretion, for such a count made close to a well might
have a very different result from a similiar count made
near the tabularium.'because of the implantation of
secondary carinae. (See Plate 1-0)

If a septum is considered as a plane, and a carina
is considered as an intersecting plane, the two structures
will be perpendicular to each other. In a longitudinal
section, which passes only through the dissepimentarium
of a corallite, the carinae will appear to be perpendicular
to the septa on those septa vhich are normal to the plane
of the section. However, the carinae on those septa which
are at an angle to the plane of the section.will no longer
appear to be normal to the septa, Reverting to geometry,
this effect is the result of passingla plane through two
mutually perpendicular planes in such fashion that the
third plane is normal to neither of the first two. When
this condition is met, the trace of the two mutually
perpendicular planes on the third plane will result in
the angles of two opposite quadrants being less than 90
degrees, while the other two angles will be greater than

(18)

90 degrees. In this case, the third plane is represented
by the plane of the longitudinal, or longitudinal-oblique
section, and the two mutually perpendicular planes are
the carina and a septum. A longitudinal-obluque section
illustrates this property to best advantage. In such a
section, the tabularium will appear as an elongate oval,
with the long axis parallel to the corallite axis. Both
immediately above and below this oval, the septa will have
been cut so that the plane of the section is normal to the
septa, and so the carinae will present their usual appear-
ance. At either side of the oval, between the tabularium
and the well, there will be septa which will have been cut
almost co-planar, and so the carinae will appear as sloping
lines on the septa. Between these two extremes there will
be septa cut at all angles from 90 degrees to zero degrees,
and the carinae on these septa will depart more and more
from.the normal noted at the 90 degree point, and will
approach the attitude seen in a section co-planar with
a septum, which is reached at the zero degree position.
(See Plate XIV-a)

Walls:-In.most Hexagonaria the walls are composed
of three layers of material; a thin, central, opaque
lamination, and a much thicker layer of translucent
material on either side of the opaque central layer (See

Plate III—a). The entire wall may vary in thickness from

(19)

PLATE II

 

CARINAE
MODIFICATIONS DUE TO ORIENTATION

 

Appearance of various forms In a section cul normal To
their slope.

 

 

—+ A A—sep’rum

lozenge modified elbow crenulate

Appearance of the same carinae in a section out at
a shallow angle lo Their slope.

 

«A A seplum

lozenge modified elbow crenulale-

 

 

CARINAE MASKED BY DISSEPIMENTS
(b) ‘

dissepiment dissepiment

+ A septum
offset yard-arms crenulates

 

 

 

 

 

 

 

 

 

 

xv! Jvill‘lir {Ur

\li. iI» l

 

20

as little as 0.2mm to as much as 1.0mm

walls range from straight or curved to strongly
serrate, or zig-zag. In many instances this variation
seems to be a function of the manner in which the septa
join the walls. If the septa of two adjacent corallites
meet the wall base-to-base, the wall.wdll.be straight or
curved, but rarely if ever serrate. However,:r the septa
are offset along the wall, the walls tend.to be serrate.
(See Plates III-b and XIV-b)

The gross appearance of serration is present in some
cases where close examination reveals the wall, as repre-
sented by its Opaque central lamina, to be straight or
somewhat sinuous. This pseudo-serration is caused by the
deposition of translucent skeletal material about the
base of each septum. In this case, if the septa meet the
wall in offset fashion, the gross appearance is that of a
strongly serrate wall. (See Plates 111-0 and XIV-b)

Septa:-Some specimens Show a dark central layer in
the base of the septa. (See Plate XIV-b) Such an occur-
snce is by no means universal. Many specimens are
utterly lacking any internal septal structures, and even
in those cases where the central lamina is present, it
may not be found throughout the corallite. Many variations
occur, from well-defined lines to shadowy areas in the base

to uniform, translucent, featureless structures.

(21)

The septa range from straight to crenulate, with all
intervening degrees of sinuosity. This is true whether
they are examined in transverse or longitudinal section.

Septal dilation:-Corals have an annual cycle of
growth, such that it produces an effect much like tree
rings. (Paul, 1942) Every structure in the coral
skeleton is thickened in cyclic fashion, but the walls
and septa are affected to a much greater degree than the
more delicate structures such as dissepiments and tabulae.
(See Plate XIV-c) The thickened portion of the septum is
not a geometrically perfect figure,‘but rather is cone
trolled and modified by local deviations. As the animal
grows upward during the period when the skeleton is being
thickened, the dilated portion conforms to the profile of
the calyx. Thus, the wall side of the thickened septa
will be higher than the axial ends, for the axial ends
were depressed into the calicular pit. (See Plates IV and
XIV-c) Such growth rings have important effects on the
appearance of transverse sections.

Should a transverse section be out such that it just
nips the peripheral portion of the thickened area, that
septum will show mural dilation. (see Plate IV) If
however, the section cuts the thickened portion through
the middle, the entire septum will'be dilated, but if the
section cuts through the axial portion of the thickening,
the septum.will be axially dilated.

Because these growth rings are cyclic, it is quite
possible to obtain a transverse-obluque section Which cuts
the thickened part on one side of the corallite, and
passes neatly between two growth rings on the other side
of the corallite. Result: septal dilation on one side
of the corallite, and none on the other. (See Plate I7)
Further, because the growths are not entirely regular,
it is common to find all combinations of these conditions
in a single corallite. Transverse serial sections out
only a few millimeters apart demonstrate that the nature
of septal dilation may change within a very small vertical,
as well as horizontal distance.

.5329. and numbers _o_f_ m:-Because of the known
relationship between size of animal and structural
organization, a comparison was made between.the cross-
sectional area of corallites and the numbers of septa
which each contained. The areas were obtained by the use
of the nomogram.on Plate VIII of this paper, and.the septa
were carefully counted and tabulated. Care was taken.so
that only mature corallites were used, and only trans-
verse sections were used so that the true cross-sectional
area would not be greatly distorted.. The result is in
cluded in Plate V'and VI).

It would seem.that there is a distinct correlation

between cross-sectional area and the number of septa.

(23)

PLATE III

 

SECTION OF WALL BETWEEN TWO CORALLITES

 

 

 

 

 

 

(0)
4"" Translucent outer layer
_Opaque central layer
¥-~ Translucent outer layer.
Magnification x30
(b) RELATION OF SEPTA TO WALL SERRATION
selpta
/' I I II \ \\‘
’l/ 0/ I; II \\ \
I / I I. \
I
well —I ' .
\i\ \\ 'I
' l I l /
II II I l l
l I 1 h
septa
Magnification xl5
(C) PSEUDO-SERRATE WALL
septa
l T

 

Opaque cores in septa.
(not universally present)

 

 

III I...
W” 'I " III I‘

l
l
septa

 

 

Translucent material.

 

 

 

Magnification x 20

 

 

 

 

 

 

 

 

 

 

 

 

24

 

 

 

This receives further oonfirmation.from a specimen
collected in Afton Quarry. (See Plate XV-a) In fliis
coral, a large corallite was forced to reduce its size.
Accordingly, a new wall was formed which is completely
separated from any other structure except two of the
original septa. In the abandoned portion.the skeleton
contains 17 septa, while in.the newly-formed section of
wall there are ten septa. The odd number of septa is
accounted for by an intricate arrangement of anestamozed
septa at one end of the new wall. Reduction in area snails
reduction in the number of septa.

.As a corollary, it seems likely that a corallite
must add additional septa as it grows larger. The process
by which this is accomplished is uncertain. However,
there is a suggestion in a specimen from the Gravel Point.
In this coral, the dissepiments‘between two widely-spaced
septa have developed a compound structure which in turn is
developing a straight central portion mid-way between the
two original septa. This implanted structure might well
be the inception of a new septum, which will eventually
extend itself to both the wall and.the tabularium. No
exception to the rule of even numbers of septa has been
found, and so it appears probable that two new septa must
be developed simulatneously, or nearly so. No trace of the
hypothetical second septum.has been noted in the specimen
described.

:\

PLATE IY

 

”GROWTH RINGS" and SEPTAL DILATION

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wall \ Profile of calyx /w0II
O A - "’__‘\ - N
\8 ~ II“ A” IIIIIII B‘ Thickened area
*“IIIIi IIIIIII .
C - 1 C
E ' \~ E'
septum septum \\

 

 

(Schematic longitudinal section

AA', 88', 00', 00', EE' = Traces of variously placed and oriented
transverse sections.

 

 

DIAGRAMMATIC APPEARANCE OF SEPTA IN TRANSVERSE

 

SECTION
mural complete axial asymmetric no
dilation dilation dilation dilation dilation
A ..... B ---- C ..... D ..... E wall

 

 

 

 

 

 

e e I

 

 

 

 

 

 

 

 

 

---_- . ————— waH

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

rvi’ I... I .
-Vlr .lIi ,

\

 

AREA orgc SEPTAL DISTRIBUTION PLATE EU.

NO 25-5 Ferron Point I

    
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

40 W W- ~ WW W W—WW 1
38 WWW W—— - , ...___.______T_ p g
I No. 244 reenshaw 2
36 _ . ———-—« m AL— 36 __..___-,l_ ___,_-_-_...__ ,
34 —WW , »— W- »4 —< — W-— WWW ‘—--Itt -_.____-._ . WIW— -—-— 34 _._...._._.__.____-_.._..____,__ ;
<1 32 --—~— — l»- W ._ w— ~—~ ,.._.-ar._._-__._.-__-__ir_._ 32
I” 30 .-4.__--__. -. -. " "v‘—""'_v"v' "v“ ‘t—v‘“ 30 I
Q. i
u; 28 _----,_-_. --l~— 28 ;
U)
26 _~-—_-~ 4r _.....— . J v -nfi—A 26
\ L 1 l 1 A l 1 1 l A L l
r r r r T r T r r .I r r 1 to O o o
.2 ‘, O O
5)) 3 ‘3 8 3 8 :3 r~ F‘ :r g on 53 9 8 :1) m m V 9 *0 53 ‘0 3
A R E A A R E A

AREA : Cross-sectional area of corallites in square millimeters.

No. I3—I Gravel Point

No. 2l-I Gravel Point

32
<30
0— 28
LLJ

a) 26

   

No. 23—2 Genshaw ;

No. 9—2 Formation unknown

   

AREA

 

 

 

 

Dissepiments:-Dissepiments appear in transverse
sections as very fine, curved traces joining adjacent
septa. They areisually a simple curved line, but this
pattern may be elaborated to include compound dissepiments
in which a second structure joins the first. (See Plates
Yea and xw:o) In longitudinal sections, the dissepiments
are seen to bound smell, globose dissepimental chambers.
If the dissepiments are simple, the chamber will be
bounded on two sides by adjacent septa, and on the other
sides by dissepiments. If the dissepiments are compound,
then the chamber is bound on only one side by a septum,
and all other sides are bound by dissepiments. These
chambers may be either horizontal or inclined, depending
on the attitude of the longest diameter as seen in longi-
tudinal section. (See Plate Veb) Some patterns are
sufficiently regular in longitudinal section to warrant
the description ”scale-like". (See Plate XVI-a) Others
are extremely diyerse. Some corallites present patterns
in which the chambers will be horizontal in one portion of
the corallite, and incline in another. (See Plate XVI-a)

Size of dissepimental chambers is defined as being
the length of the longest diameter, regardless of the
orientation of the chamber. Size is variable within wide
limits within the same corallite. One specimen includes
chambers which range in size from.0.2mm to 170mm.(See Plate
XVI-b).

(28)

If the chambers in a coral are horizontal and regularly
disposed, then a transverse section will show the dissepi-
ments as generally regular spaced, but with.the arch
alternately pointing axially and peripherally. However,
if the specimen is transverse-oblique, the same dissepi-
ments will tend to display a consistent peripherally-
pointed arch on one side of the corallite, and an axially-
pointed arch on the other side. (see Plate VII-c) Because
such regularity as is displayed in the diagram is rarely
found in nature, it follwos that the two instances cited
are the limiting cases, and only generalizations or approx-
imations of them Should be sought. Irregularities in
structure, such as variation in size or attitude can
develop many and varied patterns.

Some septa Which project into the tabularium.appear
to be connected by dissepiments rigit up to the distal
end. (See Plate XVeb) Examination in longitudinal section
discloses that these traces are in reality tabellae,
but tabellae which are nothing more than dissepiments
which terminate on a tabula rather than on another dissepi-
ment. Tabellae are probably only variations of the dissepi-
ments.

The apparent concentration of dissepiments around the
tabularium of some forms is an effect produced by the

approach of the septa to a common center. (See Plate VIII)

(29)

 

 

PLATE III

 

 

 

 

 

DISSEPIMENTS

 

wall Transverse section

 

 

 

Simple forms compound forms

 

Longitudinal section

wall wall

'9
A
(b) v"
‘7‘.
Horizontal dissepimental
chambers.

 

 

 

Inclined dissepimentai
chambers.

Arrows indicate greatest diameter of chambers. This isthe
dimension used in measurement.

 

Longitudinal section Transverse section

 

 

 

 

    

wall 8’ wall

Function of orientation as a control of direction of curvature.

 

 

(him
5 _

 

 

 

 

50

 

 

 

 

Careful measurements should be taken in the longitudinal
section to determine whether there is actually a grouping

of dissepiments around the tabularium, Such a grouping

may be produced in transverse section without decreasing

the size of the dissepimental chambers. If the chambers

in that portion of the dissepimentarium close to the wall
are horizontal, but those in the portion near the tabularium
are inclined, there will be a real concentration of dissepi-
ments about the tabularium. (See Plate XVI-a) Sanford (1939)
defines dissepimentarium.as a peripheral zone of dissepi-
ments. or course, the same effect can be produced by a
reduction in the size of the chambers.

the presence of compound dissepiments in a transverse
section is not proof positive that compound dissepimental
structures are actually present. Compound dissepiments can
be duplicated from simple ones in.transverse-oblique
sections. If the plane of the section intercepts the line
along which two simple dissepiments join, the resulting
trace will'be Yeshaped. (See Plate IX-a)

In some instances, the dissepiments are increasingly
complex near the wall. This is particularly true if the
adjacent septa are spread widely apart so that the dissepi-
mental tissue must fill a comparatively large volume of
space. These structures become so complex that the dis-
tinction between dissepiments and cystosepiments is largely
lost. Such.occurences imply that‘the two types of structure

(31)

 

PLATE YIII

 

PSEUDO-CONCENTRATION OF DISSEPlMENTS

Dissepimentarium

 

Ta bularium

All dissepiments are O.l inches apart. Note the apparent
concentration about the tabularium. '

 

 

 

 

 

 

 

 

 

 

 

 

32

 

 

 

are produced by the same mechanism in the coral animal.
Because of this relationship tabellae, dissepiments, and
cystosepiments should probably be thought of as being
members of an isomorphous series.

If the dissepiments are steeply arched, a transverse
section will usually show the traces as a broad, indistinct
line. (See Plate IX-b) Such broad, but correspondingly
faint areas are frequently overlooked When dissepiment
counts are made. The lack of dissepiments in transverse
section is not in itself proof that the fossil is free of
these structures. A transverse-oblique section will
ordinarily reveal abundant dissepiments, even then they
are not evident in transverse section. Measurements of
dissepiments Should always be made from longitudinal
sections.

Orientation Criteriat-Irom.the preceding discussion,
it is possible to develop criteria for the determination
of the approximate orientation of random.sections. It has
already been demonstrated that this is of primary importance
because of the danger of making false determinations on the
basis of a series of misinterpreted structures.

In transverse section:

l)-Carinae may be absent or few near the tabu-
larium, but relatively clear and numerous near the
wall. They will be uniformly present, although super-
ficially perhaps absent, around the entire corallite.

0—3‘ if

 

 

PLATE IX

 

DlSSEPlMENTS

 

wall ..

 

 

. dissepiment

,_--. ids—«adisseplmefii

 

M——septum

 

Apparent compound
dissepiments produced from
simple dissepiments by an oblique orientation of the section.

 

 

Orientation as a factor in the apparent thickness of
(b) dissepiments.

septum l l
septum \ ~ g ’3
\g/dissepiments 2/

 

llll'l'l’hLLLL‘

Transverse section

Pl'l'l'l’rl'fl‘lt

 

Trace of transverse section.

 

 

 

 

 

 

 

 

54

 

 

 

2)-Dissepiments will generally be simple types.
If compound forms occur, they will be distributed more
or less at random over the entire corallite, and will
not be concentrated in two opposite quadrants. (See
plate‘XVII)

3)-If there is a grouping of dissepiments, the
grouping will be consistent around the entire coral-
lite.

4)-If septal dilation is present, it will usually
be uniform around the entire corallite. Unfortunately,
irregularities in the structure of the animal may
partially invalidate this criterion.

5)-There will be approximately the same number
of dissepiments per unit of dissepimentarium on all
sides of the corallite.
In transverse-oblique section:

l)-Carinae will be abundant on one side of the
corallite, but few or absent on the other side; or,
if the carinae are short and stout on one side and
thin and tall on the other, the section will be only
slightly oblique.

2)-Dissepiments will tend to Show concentrations
of compound types in two opposite quadrants.

3)-The number of dissepiments per unit of di-
ssepimentarium may be slightly different on apposite
sides of the tabularium. This criterion is valid in

(35)

only the limited number of cases where dissepimental
chambers are either consistently inclined, or consis-
tently horizontal.

4)-If the dissepimental chambers are inclined
near the tabularium and horizontal near the wall,
a transverse-oblique section tends to eliminate the
grouping of dissepiments in one quadrant.

5)-In extreme cases, the tabularium will be
slightly oval.
In longitudinal-oblique section:

l)-The tabularium will be an elongate oval,
‘with its long axis parallel to fine walls.

2)-The carinae will be at right angles to the
septa both above and below the two apices of the
oval tabularium, but will depart from the normal more
and more as the short axis of the tabularium is
approached.

3)-Carinae will be relatively abundant on the
septa beyond one apex of the tabularium oval, but

relatively rare beyond the other apex.
AREA NGMOGRAM

Paleontologists have been accustomed to measure the
size of Hexagonaria by means of a relation called average
diameter. The quantity is defined.as the arithmetic mean
of thelbngest and shortest dimension, taken roughly at

(36)

right angles to each other. (Paul, 1942)

Biologists have long known that the morphology of
animals is at least in part a function of the mass of the
animal, because it is the mass Which controls such things
as the strength of the structural members. The mass of
animals is directly proportional to the volume, and so it
follows that any attempted correlation between morpholog-
ical characteristics and size must be made on the basis of
volumetric considerations. Fortunately, the colonial coral
is essentially a tubular animal, and for this reason the
volume of the animal is directly proportional to the cross-
sectional area.

The average diameter is not proportional to the cross-
secticnal area of Hexagonaria. If a square two inches on
a side is measured, using average diameter, the resultant
average is two inches. If a rectangle four inches long
and one inch wide is measured in the same way, the resulting
average diameter is two and one-half inches. And yet the
areas of both figures are the same! Further, if our
rectangles were the tops of solid blocks four inches high,
the volume would in each case remain the same. Average
diameter is not proportional to volume.

These considerations make it obvious that average
diameter is not a valid measurement of size. A direct

measurement of cross-sectional area would be ideal,

(37)

because the volume of a coral is directly proportional
to its cross-section. Hewever, measuring areas directly
with a planimeter is exceedingly laborious, and so an
easier method was sought. A nomogram (Plate X) was
developed Which yields rather close approximations to
the true values. It involves measuring two dimensions
at right angles to each other, and such that the lines
bisect each other. The greatest length is chosen for
one, and the other is allowed to fall where it may. The
method is based on the observation that most Hexagonaria
are irregular hexagons. It is obvious that the results
achieved by the nomogram.are only approximations, but it
is believed that they are sufficiently accurate to warrant

future usage.
CONCLUSIONS

Amond the specific criteria used to identify Egggr
gonaria are the following: the character of the well,
whether straight or serrate; the shape, number, and mode
of occurence of dissepiments; the average diameter of the
corallites; the number and nature of the carinae; and the
number and nature of the septa.

The interpretation of these structures with respect
to solid geometry has been somewhat neglected. Geometric

considerations indicate that the shape of the carinae, as

(38)

 

 

_ 0.2

 

L---O.l

 

Nomogram for the area A of a semi-regular hexagon
in square units, in terms of ”Height” H and "Length” L.

Method of use:

l)— First index line from H to L determines K.

2)-— Fmd K2 : Kl

3)’ Second index line from K2 to H determines area A

 

 

 

 

 

 

a a T
H
0 O
o l
n L K,
no.0
E
t.
L L
E430 . 50
F 7
P20 P40
i.
g... -
~——lO / —3.o
: /////
:/~/""
-,/
// ““5 .._2.0
_ 4 _
L3 ;,5
L
l—2 :
i
l—l.O
i
——| P///
i‘
ii
»—o.5
0.4
i- 0.3
~ 0.2
—0.1

 

 

o:
O\

l
A
O

l
o:
O

1 1‘7111

b-
9-20

_
r—lO
,.

 

PLATE 2:

 

 

 

 

 

 

 

 

 

 

well as their numbers, are considerably modified by the
orientation of the plane of the section being studied.

The same is true of dissepiments, for their direction
of curvature, their number and spacing may be altered by
a comparatively slight change in the orientation of the
study section.

In many cases the dilation of septa appears to have
little value, because the dilation is controlled by the
orientation of the section.with respect to growth lines,
and perhaps any other structural modification which results
in a thickening of the skeletal structures. Certainly
septal dilation as a specific criterion should be used with
caution.

It appears that the average number of septa has little
value in specific determination, because of the correlation
between the size of the corallites and the numbers of septa.
It is easily possible, for example, that an entire colony
might be stunted by turbid water, and thus display a con-
sistently lower number of septa per corallite than a more
advantageously located colony a Short distance away.

The nature of the wall is of doubtful value because
of the control exercised by the septa, and also because of
the development of pseudo-serrate walls, which serves to

mask the true nature of the structure.

(40)

Most identifications of Hexagonaria have been based
upon the examination of sections the orientation of which
have been assumed to be identical, 1. e., perpendicular
to the axis or parallel to the axis. This laboratory
study applying defined orientations of sections suggests
that a re-examination of established species would be

desirable.

(41)

BIBLIOGRAPHY

Darrah, W. 0., (1936) T13 Peel Method in Paleobotany,
Harvard Botanical Museum Leaflets, Vol. 4,
No. 5

Cooper, G. A., See Warthin, A. S.

Paul, H., (1942) Prismatophyllum in the Traverse Group
of North-central Michigan, M. S. Thesis, Michigan
State College

Fenton, M. A., (1935) Nitrocellulose Sections of Fossils
and Becks, American Midland Naturalist, Vol. 16,
pp. 410-412

Grabau, A. W., (1913) Principles _q__f Stratigraphy, A. G.
Weiler, New Yoik‘

Kelly, W. A.. ( ) Lithostrotiontidae of the Rocky
Mountains, Journal 9;; Paleontology, Vol. 16,
pp. 351-361

Kelly, W. A. and Smith, G. W., (1947) Stratigraphy and
Structure of the Traverse Group in the Afton-
Onaway Area, Michign, American Association of
Petroleum Geologists BuIIetIn, No. 3, 701. 31',-
March

Ma, T. Y. H., (1934) On the Seasonal Change of Growth
in a Reef Coral, Favia Speciosa (Dana), and the
Water Temperature of the Japanese Seas During

the Latest Geological Times In erial Academy
9; Japan, Proc. Vol. 10, pp: 353-3

(1937) On the Growth Rate of Beef Corals and
its Relation to Water Temperature, National

Institute of Zoologyan dBctan6 (Academia
SEIca) Menu, 6pages

Sanford, w. c.. (1939) A Review of the Families of Tetra-
corals, American Journal of Science, Vol. 237,

PP. 295-523, PP. .-

(42)

Simpson, G. G., (1937) Patterns of Phyletic Evolution,
Bulletinhgg. the Geological Society 9; AmericaI
701. E, pp. 503-314

Slcss, L. L., (1939) Devonian Rugose Corals of the
Traverse Beds of Michigan, Journal _o_f_ Pal eon-
tologz, Vol. 13, pp. 52-73

Smith, G. W., (1942) The Ferrcn Point and Genshaw
Formations in Cheboygan and Wests rn Presque
Isle Counties, Michigan, M. S. Thesis, Michigan
State Colleg

Stumm, E. C., (1937) The Lower Middle Devonian Tetra-
corals of tie Nevada Limestone, Journal 9;
Paleontolog, Vol. 11, pp. 423-443

(1938) Upper Middle Devonian Rugose Corals of
the Nevada Lime stone , Journal gt; Paleontolog,
V01. 12, PP. 4.78-485

Swann, D. H., (1947) The Favosites Alpenensis Lineage
in the Middle Devonian Traverse Group of Michi-

gan, Cont. Museum of Paleontolo , Univ. of
Michigan, 701. 'VIf'Krt. 9

Vaugian, T. W., (1917) Corals and the Formation of Coral
Reefs, AnnualReport _o_i; the Smithsonian Institute,

Verhoeven, C. S., (1948) Possibility of Determining
Restricted Reef Conditions from the Areal Dis-
tribution of the Widely Distributed Contempo-
raneous Detrital and Lagunal Facies, M. S.
Thesis, Michigan State Colleg

Warthin, A. S. and Cooper, G. A., (1934) Devonian Studies
in Southwestern Ontario and Mich igan,Ex lorat ion
andFieldWork 9; the Smithsonian Inst tute,

pp. m

(43)

APPENDIX

Derivation of Area Nomogram:
a.

a q j
H

Q Q
a l

L .—

 

 

 

Let L equal length of a semi-regular hexagon, and
let H equal the height of such a hexagon, and
let a equal the lengths of the sides.

ThenL=a+m .

Solving for a=’(W-L),

then H(a+L)=A: may ALZ+3H ).
Setting L= KH, then A: gist—TT?)

Writing both in logarithmic forms, and then in determinate
forms, which suggests an alignment chart:

2 log H l
3 log 6A = O
4 E iogtzK+ 4K +3) 1

O lch l
l i: log L 1
2 log H l

:0,

 

 

 

Let H \ary from 1 to 16, and L from 1 to 100, then

X,’0, y,: logK; K: (- l, 100); :(-l, 2).
12:1: y,_= $10311; L:(1:100): y “(0 1)-
x,=2, y:: logH; H:(1, 10); y: (C,

1.)
x.,=3, log6A; 32 666); y :(.07, 1.15)
x,=4,=103(2K+r4K‘+3'; .,1 100); y:(.l4, 1.3).

But this shape is inconvenient, and a chart with better
groportions results from the following modifications:

0 3logK 3 1 7 9logH 1
1.75 31081. g 1 = o, 5.25 giogeA - .65 1 =0
7 4 7 3.5 4
9logH 1 _9_log(2K+ fix’+3')-1 . s 1
q,

 

(44)

The new scales are:

x,—-o, y,:SlogK+3; K:(.l, 100); y :(o, 9).
x,=l.75, yzzglogL+_9_; L:(.l, 1000); y :(0, 9).
4 ‘I

x,=7, y,=9logH; H:(1, 10); y :(0, 9).
14:5.25, m:%log6A-.65; A:(.32, 667); y :(0, 9.71).

x,=3.5, yrzgngmK-thK‘t' 2.5-1.3; K:(l, 100); y :(0, 10.42).
Hence the chart will occupy a space 7 inches wide and A
10.5 inches tall. With the range of the scales suitably

modified, the resultant chart will be found on page 39
of this paper.

(45)

PLATE XI

ORIENTATION OF‘SECTIONS (a)

The white planes in the figure indicate the
planes of diamond-saw cuts used to obtain oriented
and serial sections.

TYPES OF CARINAE (b)

Yard-arms~

-—Lozenges
Elbows

 

Crenulatel

(46)

TYPES OF CARINAE

Sigmoid
No. 550

Arched
No. 15

EXTENDED CARINAE

PLATE XII

(a)

Bow-shaped
No. 200

(13)

Note trace of carinae
extending beyond the axial
terminus of the septa.

No. 550

(4'7)

PLATE XIII

IMPLANTED CARINAE ' (a)

 

No. 300 QINo. 14638
CARINAE IN TRANSVERSE-OBLIQUE SECTION (b)
No. 14638

MASKED CARINAE (c)

No. 450

IAa\

PLATE XIV

CARINAE IN LONGITUDINAL-OBLIQUE SECTION (a)
No. 200
No. 550
No. 550
WALLS (b)

Only a very faint
trace of the laminated
wall structure is
present in the photo-

graph.

No. 35

GROWTH RINGS' (c)

No. 7-6 (After Faul, 1942

(49')

AREA-SEPTA CORRELATION

No. 100

DISSEPDJENTS AND TABELIAE

COMPOUND AND SIMPLE DISSEPIMENTS

compound~

NO. 7.6

No. 35

(50)

PLATE XV

(a)

(b)

(6)

#simple

PLATE XVI

DISSEPIMENTAL CHAMBERS (a)

No 450

Chambers peripherally horizontal, axially inclined.
Pattern regular, or scale-like.

SIZE RANGE OF DISSEPIMENTAL CHAMBERS (b)

No. 100 No. 15

Note that the size range of the dissepimental
chambers is very great in these specimens.

(51)

PLATE XVII

DISSEPIMENTS IN TRANSVERSE-OBLIQUE SECTION

\\simple dissepiments
in these quadrants

- compound dissepiments
in these quadrants

No. 450

Note the tendency of the compound dissepi-
ment s to concentrate in two opposite quadrants.
This is a criterion.tc>establishing the section
as transverse-oblique.

(52)

PLATE XVIII

SERIAL SE CTI CNS

Spec imen No . 450

(53)

"in . .
f «‘9‘;
'I: r‘t

Oa-fl! I "A

U "'l’
I

Q

l

‘r
l ‘ '

' ‘ i

\

l

4
l
8

Remit um um.)