HEAVY MINERAL INVESTIGATIONS OF SOME FODSOL SOIL PROFILES
IN MICHIGAN

By
HOY FETER MATELSKI

A THESIS

Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and Applied Science
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

DEPARTMENT OF SOIL SCIENCE

1947

P roQ uest Num ber: 10008376

All rights reserved
INFO RM ATIO N TO ALL USERS
The quality o f this reproduction is dependent upon the quality of the copy subm itted.
In the unlikely event that the author did not send a com plete m anuscript
and there are m issing pages, these will be noted. Also, if m aterial had to be removed,
a note will indicate the deletion.

uest
ProQ uest 10008376
Published by ProQ uest LLC (2016). Copyright of the Dissertation is held by the Author.
All rights reserved.
This w ork is protected against unauthorized copying under Title 17, United States Code
M icroform Edition © ProQ uest LLC.
ProQ uest LLC.
789 East Eisenhow er Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6

ACKNQWLEDGMENTS
The author is grateful to Dr. L. M. Turk and
Prof. J, 0* Veatch for assistance, advice, and encourage­
ment in the research reported in this manuscript.

To

Dr, Bennett T. Sandefur of the G-eology Department, he
is indebted for guidance in the petrographic study.
To the following staff members of the Soils Department
the author wishes to express appreciation and thankss
to Dr. C. E. Millar who instituted and guided the early
stages of the problem; to Dr. N. S, Hall and Mr. A. H. Mick
who generously contributed to the more technical phases of
the work; to Dr. R. L. Cook and Mr. Lynn S. Robertson, Jr.
who assisted in the preparation of the photomicrographs.

190282

CONTENTS
Page
Introduction. .................................. 1
Review of Literature.........

1

Collection and Sampling........

6

Description of Soil Types..........

6

Preliminary Studies

...........

11

Determination of Reaction, Free Alumina,
Iron Oxide, and Colloidal Silica....... 11
Mechanical Analyses
Heavy Mineral Studies

..........
.............

Separation of the Heavy Minerals.......

12
13
13

Mounting of the Heavy Minerals.......... 14
Distribution of the Total Heavy Minerals 14
Microscopic Identification and Counting
of Mineral Crains
........

15

Description of Hee„vy Minerals........... 16
Presentation of Mineral Count Data...... 19
Discussion of Results..............
Neubauer Tests

........

Summary and Conclusions.....

..........

Bibliography...,...........

21
28
31
33

APPENDICES
Tables 1 to 9.......................

37

Plates 1 to 5 ...............................

47

Figures 1 to 17...........

52

INTRODUCTION
In a study of the development of podsols it is desirable to in­
clude an investigation of the heavy minerals.

Very little informa­

tion can be found on the heavy minerals of podsols in the United
States; none can be found on Michigan podsols.

To recognize geo­

logical differences in soil horizons, to determine the intensity of
the weathering processes in various horizons, and to account for the
Observational differences in the soil profiles have been major ob­
stacles to the student in the study of soil genesis and development.
Recent improvements in the preparation of soils for petrographic
analysis and newer techniques in the separation and microscopic
identification of soil minerals have aided in more quantitatively
determining the changes that may have progressed within soil profiles.
This investigation was an outgrowth of observations on soil
profiles made while conducting a land type survey in Charlevoix and
Presque Isle Counties of Michigan.

Several theories were advanced

to explain the profile differences in the various soils.

It was

believed that through the utilization of the recent techniques in
the study of heavy minerals and through an analysis of the data
supplied by these techniques certain morphological pecularities
observed in some Michigan podsol profiles might be explained.
REVIEW OF LITERATURE

The study of the heavy minerals in podsols has not been very
extensive.

This has been due in part to the lack of suitable

techniques for the separation and positive identification of the

2.
heavy minerals.

Recently more simplified petrographic methods have

enabled the less highly trained petrographer to identify minerals
in soils,
Cady (3) differentiated the mineralogical characteristics of
podsols from brown podsolic soils.

The heavy minerals in the A

horizon of podsols were decidedly less than those in the C horizon.
The minerals hornblende and hypersthene weathered rapidly while
magnetite and garnet were little affected.

In the A and C hori­

zons of the brown podsolic profiles, the heavy minerals were found
in similar amounts,
Richard and Chandler (33) investigated three strongly developed
podsol profiles from Quebec Province, Canada and found that horn­
blende and to some extents hypersthene weathered rapidly in the Ag
horizon whereas it was only slightly weathered in the C horizon,
Jeffries and White (22) showed that in the Leetonia sand
(podsol) zircon and anatase were abundant and typical; tourmaline,
rutile, muscovite, chlorite, epidote, barite, magnetite, and
leucoxene were present but not in sufficient amounts to be typical.
The most recent procedures for the isolation and microscopic
identification of the heavy minerals have been outlined by Marshall
and Jeffries (28).

These include methods for the removal of in-

crusting substances and oxide coatings, procedures for the sepa­
ration of the soil minerals, and aids for mounting and counting
the separated soil minerals.
Many workers, Cady (3), Haseman and Marshall (16), Humbert

3.
and Marshall (17), Marshall and Jeffries (28), Marshall (27),
McCaughey and Fry (29), Mickelson (30), and Richard and Chandler (33)
have investigated the use of the resistant heavy minerals such as
zircon, tourmaline, garnet, anatase, rutile, and magnetite as indi­
ces of soil maturity and development*
Haseman and Marshall (16) have reviewed thoroughly the various
minerals resistant to conditions of weathering.

Zircon if present

in sufficient quantities seems to he the ideal immobile indicator
for measuring changes in profile development*
Dryden and Dryden (9) in comparing the resistance to weath­
ering of fresh and weathered Wissahickon schist from Pennsylvania
and Maryland found zircon and green hornblende to be more resistant
than garnet.

Goldich (13) has reported opposite results on an

amphibolite from the Black Hills; garnet was more resistant than
hornblende.

This may be attributed to the differences in climate,

slope, or vegetation.
Other workers have stressed the light minerals in explaining
soil formation and development.

Fieger and Hammond (ll) studied

the effect of cultivating rice and flooding with fresh well water
on the minerals of some coastal prairie soils of Louisiana.

They

found that limonite, hematite, and magnetite are largely lost from
the silt fractions of the A, B, and C horizons of the flooded soil.
The feldspars, particularly the potassium feldspar, were more stable
then the iron oxide minerals.
Jeffries and White (20, 21, 22, 23) investigated podsols, gray

4.
brown forest soils, and brown forest soils and found that the
influence of parent material is of outstanding importance in the
development of soils.

The minerals of limestone soils from Indiana,

Virginia and Pennsylvania were characterized by high amounts of
feldspars in the very fine sand fraction.

In the Hagerstown series

microcline appeared to be the resistant mineral.
Marshall and Jeffries (28) believe that in addition to the
heavy minerals certain feldspathic minerals and muscovite may be
useful in weathering studies.

Microcline, a resistant mineral in

some soils, may be utilized to measure changes in profile development.
Jeffries (19) has discussed the recent advances in soil miner­
alogy.

Among these is the use of the double~varis.tion method as

a means of determining the exact chemical composition of the soil
minerals.

If this technique becomes simplified, it may be possible

to measure changes in the chemical composition of soil minerals
due to weathering.
The methods of determining the quantities of minerals present
in a particular fraction have been investigated by many pedologists
(3, 12, 16, 18, 28) and petrographers (2, 4, 5, 25, 31).

The

method generally used is to separate the sand into grade sizes
and count the number of particles of each mineral.

Then with the

following formula (12), the volume is calculated:
Percent by volume s Number of particles of one constituent x 100
Total number of particles counted
Chayes (8) determined the average grain weights of sized parti­
cles by counting the number of grains per milligram.

5.
The errors involved in heavy mineral studies have been eval­
uated by Krumbein and Rasmussen (26).

They found that after sub­

jecting 24 closely spaced samples of beach sand to heavy mineral
analyses the sampling error was about 10 percent, and the errors
due to splitting, separating, and counting were of approximately
the same order of magnitude.
Rittenhouse (34) from a study of more than 20 species of
heavy minerals has calculated the probable errors due to counting.
These probable errors are expressed as percent of the heavy min­
eral’s frequency and as percent of the total number of all heavy
mineral grains.

For example, a counting of 400 grains with a 3

percent frequency would have a probable error of 19.2 percent of
3 percent.
Chayes (7)

found that the use of the number of grains counted

as an index of the counting error required experimental evaluation.
In Chayes * studies the critical range for number frequency analy­
sis was between 500 and 2000 particles.

Often the increased accu­

racy of counting 1C00 grains instead of 500 grains is significant.
On the other hand, Pye (32) counted 25, 50, 75, 100, 200, and 300
grains from random fields on slides representative of beach sands,
sandstone outcrops, and oilwell cores.

From his studies 50 grains

appeared to be the optimum number to count.
From this review of literature it would seem that an investi­
gation of the heavy minerals in podsols would provide valuable and
needed information in the recognition of various podsols and a basis
for the measurement of the changes within the profile.

COLLECTION AND SAMPLING

Soil profile samples were collected under as near virgin forest
conditions as possible.
the profiles.

A large L-shaped pit was dug for each of

Horizontal variations were taken by sampling along

both arms of the "L” ; large samples approximating 50 pounos were
taken of each horizon.

DESCRIPTION OF SOIL TYPES
Two groups oi podsol soils were investigated, namely:

Emmet,

Kalkaska, and ¥/allace sands which were tentatively segregated as
Group I; and Grayling, Rubicon, Roselawn, and Eastport sands here­
after designated as Group II.

Group I is characterized by a

stronger expression of morphological features of typical podsol
profiles than is Group II.

The Kalkaska and Emmet sands supported

a forest cover of hardwoods, whereas the other types supported a
pine cover.

Veatch (41) estimates the total area in Michigan of

the soils in Group I to be 1,350,000 acres5 in Group II, 4,200,000
acres.

Where an asterisk appears in the following profile descrip­

tions, a detailed investigation was made of the heavy minerals.
GROUP I. Emmet sand
Emmet sand differs from Kalkaska sand in that it occupies
the sand hills, whereas Kalkaska sand occurs on the nearly level
sand plains and valley floors.

The typical profile where sampled

(NWj- sec.8,T.35N.,R.4E., Presque Isle County) consisted of:
(l)

a thin

to 1-inch litter layer, composed chiefly of

partly disintegrated leaves, grasses and wood.
*(2)

a 2- to 3-inch dark-brown to nearly black, slightly
acid medium sand layer containing many partly decayed
roots and leaves.

(3)

a 6- to 8-inch light-gray, slightly acid, leached, loose
medium sand layer.

*(4)

a 6- to 8-inch umber-brown or dark-brown, slightly acid,
weakly cemented medium sand layer.

(5)

a 16- to 20-inch grayish-yellow or yellow, neutral me­
dium sand layer.

*(6)

a grayish-yellow, strongly alkaline medium sand ex­
tending to a depth of 60 inches.

Kalkaska sand
This type occurs on level, dry, hardwood plains and valley
floors.

This series was distinguished from the Rubicon, Grayling,

and Eastport by its umber-brown sand layer which underlies the ashgray surface soil and by the greater amount of cementation in the
accumulation horizon.

The profile, sampled in NE-|: sec .8,T.32N. ,

R.5W.,Charlevoix County, consisted of:
*d)

a 2- to 3-inch surface layer of dark-brown to nearly
black, slightly acid, medium sand mixed with partly
disintegrated leaves, twigs, and pieces of wood.

(2)

a 8- to 10-inch light-gray, strongly acid, loose,
leached medium sand layer.

*(3)

a 2- to 3-inch umber-brown or dark-brown, strongly
acid, weakly cemented medium sand layer.

(4)

a 4- to 6-inch lighter colored dark-brown, slightly
acid, weakly cemented medium sand layer.

*(5)

a loose, pale-yellow or gray, slightly acid, medium
sand to 60 inches.

V/allac e sand
Wallace sand comprises the
sent old

dunes or

low pineland ridges which repre­

beach ridges.It differs from the other dry

sands sampled in its greater thickness and stronger cementation
of the brown horizon.

A description of this soil type, collected

in NE-J sec .3,T.33N. ,R.5E., Presque Isle County, follows:
(l)

a -J- to -J— inch dark-brown, slightly acid mat of disinte­
grated grasses, twigs and pieces of wood.

*(2)

a 2- to 3-inch layer of dark-grayish brown, slightly
acid sand mixed with partly decomposed plant roots.

(3)
*(4 )

a 6- to 8-inch loose, ash-gray, medium acid sand layer.
a 10- to 14-inch dark-brown, medium acid sand layer,
cemented in places into

a hardpan.

Part of the sample

included dark-brown fingers which projected

to a depth

of approximately 3 feet into the lower horizons.
(5)

a 18- to 20-inch lighter colored dark-brown, weakly
cemented, slightly acid sand.

*(6)

a loose, brownish-yellow, slightly acid sand layer
sangpled to a depth of 70 inches.

GROUP II. Grayling sand
The soils of the Grayling series occur as the very dry, level,
sand plains which originally supported a pine vegetation.

The

present growth consists chiefly of thickets of jack pine, together
with sweetfern and bracken.

The profile, sampled in SS^r sec.7,

T.25N. ,R.3Y,r., Crawford County, did not have a definite leached
horizon but included the following layers:
(l)

a ■£- to ^-inch dark-brown organic layer, composed chiefly
of disintegrated pine needles and grasses.

*(2)

a 3- to 4-inch gray-brown, slightly acid, medium sand
layer containing some decaying plant roots.

*(3)

a 20- to 24-inch loose, brownish-yellow, strongly acid,
medium sand layer.

*(4)

a pale-yellow, slightly acid, medium sand to a depth of
50 inches.

Rubicon sand
Rubicon sand also occurs on dry, level to undulating plains.
Originally the growth consisted chiefly of white and Norway pine.
The second growth is mostly jack pine and aspen with large open
areas of bracken, sweetfern, and blueberries.

This soil supported

a little heavier growth of vegetation and has a thicker gray and
brown subsurface layer than the Grayling and Eastport sands.

The

profile, observed in NW4. sec .33 ,T.36N. ,R. 2E., Presque Isle County,
consisted of:
(l)

a -5- to -g—inch litter layer composed chiefly of disinte­
grated pine needles and twigs.

*(2)

a 3- to 4-inch dark grayish-brown, very strongly acid,
medium send mixed with charred organic matter and plant
roots,

10.
(3)

a 4- to 6-inch ash-gray, leached, very strongly acid,
medium sand layer.

*(4)

a 12- to 18-inch brownish-yellow, slightly acid, medium
sand very slightly cemented and compacted.

*(5)

a loose, pale-yellow or grayish-yellow, neutral, medium
sand extending to a depth of 60 inches.

Roselawn sand
Roselawn sand is found on rolling to hilly terrain.
profile, collected in

The

sec.29,T.36N.,R.2E., Presque Isle

County, consisted of the following layers:
(l)

a thin -g- to 1-inch layer of acid litter composed
chiefly of twigs and pine needles.

*(2)

a 2- to 3-inch loose mixture of dark-gray, neutral,
medium sand, charred organic matter, and plant roots.

(3)

a 2- to 3-inch layer of leached, gray, or light brownishgray, incoherent, slightly acid, medium sand.

*(4)

a 6- to 8-inch yellowish-brown, slightly acid, medium
sand layer very slightly cemented and compacted.

(5)

a 16- to 18-inch layer of yellowish, loose, slightly
acid, medium sand.

*(6)

a pale-yellow, loose, very slightly acid, medium sand
to a depth of 6 feet.

At 6 feet the sand appears slightly

coarser.
Eastport sand
Eastport sand occurs largely at the margin of Lake Michigan
and Lake Huron on the nearly level to wavy plains and benches.

11.
The profile is weakly developed and does not show as much yellow
coloration as the other soils studied.

The profile where sampled

(SE-^r sec .6,T.35N. ,R.5E., Presque Isle County) consisted of:
(l)

a

to -J-inch dark-gray organic layer comprised chiefly

of disintegrated pine needles and grasses.
*(2)

a 2- to 3-inch layer of gray-brown, slightly acid medium
sand mixed with many partly decomposed plant roots.

(3)
*(4)

a 3- to 4-inch layer of gray, slightly acid medium sand.
a 13- to 15-inch pale-yellow, loose, slightly acid medium
sand.

*(5)

a grayish, loose, slightly acid medium sand layer sampled
to a depth of 50 inches.

Hereafter, in the discussion of the soil types, the first layer
of each soil profile subjected to a detailed investigation of the
heavy minerals will be known as the A horizon; the second, the B
horizon; the third, the C horizon.
PRELIMINARY STUDIES

After the field samples were air-dried, they were passed
through a 2 mm. sieve.

The aggregates remaining on the sieve

were gently crushed and again passed through the sieve.
remaining coarse material was discarded.

The

The less than 2 mm.

soil material was mixed by rolling on clean paper and quartered
to approximately 20 gm. samples.
Determination of Reaction. Free Alumina. Iron Oxide, and Colloidal
Silica
The field soil reactions were taken as outlined by Spurway (361!.

12*
The pH of the soil was determined in the laboratory with the glass
electrode*

The ratio of soil to water by weight was 1:1*

The

mechanical analyses, pH, and organic matter content of the soils
are given in Table 2.

The pH values of the B horizons are, for

the most part, lower than the values for the corresponding C
horizons.
The organic matter, free alumina, iron oxide, and colloidal
silica were removed as outlined by Truog et al (40).

Marshall and

Jeffries (28) believe that of the three most commonly used pro­
cedures (16, 18, 40), the HC1 method (16) more seriously attacks
the minerals soluble in acids than Truog*s (40) sodium sulphideoxalic acid method or the Jeffries* (18) aluminum-oxalic acid
method.

However, since aluminum was to be quantitatively deter­

mined in the B horizons, Jeffries* method was abandoned even
though it least attacks the minerals.
The methods used for the determination of the free alumina,
iron oxide, and colloida.1 silica were those employed by Robinson (35)
and Truog et al (40).
It might be mentioned in passing that the results of this
analysis (Table l) indicate that there was a greater coating of
free iron oxide around the particles in the Group I than in the
Group II soils.

The presence of this greater coating of iron

oxide will be explained later in the discussion of "Opaque Minerals".
Mechanical Analyses
The mechanical analyses were determined by the method as out­
lined by Truog et al (40).

The fine sand (0.25-0.02 mm.) was

13.
obtained by sieving the total sands (2-0.02 mm.).

Sieving was

accomplished by shaking the sand on an 8 cm. number 60 standard
sieve with a vertical shake-type apparatus manufactured by the
American Instrument Company, Washington, D.C.
"time was twenty minutes.

The average shaking

Twenty minutes is usually considered

too long to shake sand particles, but recent work of Swineford
and Swineford (37) showed that the use of certain mechanical
shakers for a period of two hours did not produce a considerable
degree of wear or breakage of sand grains.
The mechanical analyses, pH, and organic matter content of
the soils are given in Table 2.
With the exception of the organic matter in fhe A and B
horizons these soils consist largely of sand.

The percentage

of silt, clay, and organic matter is greater in Group I than in
the Group II soils.
The results of the mechanical analyses indicate a uniform
parent material.

A separation of the sand into small size

fractions would assist in more definitely determining the uni­
formity of the parent material.

HEAVY MINERAL STUDIES

Separation of the Heavy Minerals
Heavy mineral separations were made on the fine sands
(0.25-0.02 mm.).

The heavy liquid used for the separation was

8, mixture of s. tetrabromoethane and nitrobenzene having a spe­
cific gravity of 2.94.

Specific gravities were controlled with

14.
a pycnometer.
One to 2-gram samples of fine sand (0.25-0.02 mm.) were
placed in small centrifuge tubes which contained the mixture of
s. tetrabromoethane and nitrobenzene.
at approximately 750 r.p.m.

These were centrifuged

At 15 minute intervals the tubes

were removed, stirred vigorously,, and replaced in the centrifuge.
After approximately an hour of centrifuging, all the heavy min­
erals had settled in the tubes.
decanted.

The light minerals were then

The particles clinging to the centrifuge tubes were

washed with the mixture of s. tetrabromoethane and nitrobenzene
and recentrifuged.

The remaining light minerals were decanted

and the heavy minerals washed into a beaker with acetone.

After

being washed twice with additional acetone, the heavy minerals
were dried at 70 degrees Centigrade and weighed.

Duplicate

samples were analyzed.
Mounting of the Heavy Minerals
After the total heavy minerals were weighed, they were
spread on a gelatinized slide.

This slide was prepared similar

to the method described by Fairbairn (10) and Marshall and
Jeffries (28).

The use of the gelatin coated slides permitted

the changing of immersion liquids without losing the grain.

No

difficulty was experienced in obtaining the Becke tests for
refractive indices.
Distribution of the Total Heavy Minerals
The percentage of total heavy minerals in oven dry, organic
free soil is given in Figure 1.

The results show that there is

15.
a significant increase of heavy minerals in all horizons of Group I
(Kalkaska, Emmet, and Wallace) sands over those in Group II (Grayling,
Rubicon, Roselawn, and Ea.stport) sands.
In all the soils, the total amount of heavy minerals was great­
est in the C horizon.

In the majority of the soils the B horizon

had the least amount of heavy minerals.

In the C horizon the a-

mount of heavy minerals in Group I is approximately twice that of
Group II.
The most striking point brought out by the studies is the
grea.t difference in the <§uantity of total heavy minerals between
these two groups of soils.
Photomicrographs of the heavy minerals of the C horizons
are shown in Plate 4.
altered.

These heavy minerals were only slightly

In Plate 5 the characteristic well-rounded grains of

the Wallace sand are readily discernible.

No other soil studied

had the mineral grains as well-rounded as those in the Wallace
sand.

Since the sand particles in the C horizon as well as the

A horizon were similarly rounded, it is believed that the parti­
cles were rounded by water prior to deposition.
Microscopic Identification and Counting of Mineral Grains
After the heavy minerals were mounted on gelatinized glass
slides, they were identified, sized, and counted.

Before the

identification of the soil minerals was undertaken ms.ny known
detrital mineral mounts were closely studied.

Frequent reference

was made to the standard texts of Johannsen (24), Milner (31),
and Krumbein and Pettijohn (25).

Identification of the opaque

minerals was facilitated by a magnetic separation. * The heavy
minerals were spread out on a flat surface and a horseshoe magnet
was passed above the minerals.

A piece of thin paper placed over

the poles assisted in releasing the magnetized minerals.

A pre­

liminary check showed that the horseshoe magnet used would remove
magnetite but not ilmenite.

Identification of all the heavy

minerals was primarily limited to determining the approximate
refractive index, pleochroism, birefringence, extinction angle,
color, and shape.
A grid micrometer ocular was used to measure the size of the
grain.

The maximum cross-sectional area of each grain was deter­

mined as follows:
i A i B iC i
* t

r~

*

The hornblende particle is illustrated in black.

The three

grid squares are shown below the letters A, B, and C.

To obtain

the maximum cross-sectional area of the hornblende particle the
grid squares that equal the hornblende particle size are counted.
In the above figure the maximum cross-sectional area is "3,f.
Description of Heavy Minerals
Hornblende
Hornblende as it occurred in the materials studied is charac­
terized by elongated prismatic grains, distinct prismatic cleavage,
ragged ends, green colors, and weak pleochroism.

It occurs as

green grains in many shades, dark and gray-green being prevalent

17.
although a few light green and some brown varities were also found.
The color density of the grains usually increased from the margin
to the center.

Well developed striae were present in some of the

grains of the parent material.
Epidote
Epidote was distinguished by its high refractive index and
strong birefringence.

It occurred as light green or bottle-green,

angular to partly rounded grains.
chroic.

Usually it was strongly pleo-

No inclusions were observed.

Garnet
Garnet was distinguished by its isotropism, high relief, and
high refractive index.
colorless grains.

It occurred primarily as pink, brown, and

A few inclusions of quartz were observed.

The

grains were mostly rounded although some angular irregular grains
were present.

Many were etched and pitted.

Garnet was distin­

guished from spinel by its index, conchoidal to irregular fracture,
and absence of cleavage.

Most of the garnet examined was isotropic

although a few grains showed anomalous strain features and weak
anisotropism.
Opaque Minerals
Magnetite represented approximately 90 percent of the total
of all the opaque minerals examined.

Of the remainder ilmenite

was dominant•
Magnetite occurred as irregular to rounded grains which were
a bright black in reflected light and opaque in transmitted light.

A few grains exhibited small shiny facets.

This mineral was

k
18.

♦

difficult to distinguish from ilmenite.

The higher magnetism of the

magnetite was sometimes of assistance.
Ilmenite occurred as iron-black to steel-gray irregular subangular grains in reflected light.
mitted light.

The grains were opaque in trans­

It was generally distinguished from magnetite by its

rhombohedral form, conchoidal fracture, more metallic luster, and
weaker magnetic properties.
Hematite represented less than 1 percent of the opaque group
of heavy minerals*
grains.

It occurred as reddish-brown irregular earthy

The hematite observed in these studies might have been

thin films enveloping the magnetite grains.
Leucoxene occurred as white to yellowish-white earthy trans­
lucent to opaque rounded aggregates.
Limonite occurred as rounded powdery granules.
in reflected light and opaque in transmitted light.

It was brown
Only traces

of limonite were found.
Minor Minerals
The following descriptions are of heavy minerals occurring in
small amounts, comprising usually less than 1 percent of the total
grains counted.
Zircon was characterized by a very high refractive index,
high relief, high order interference colors, and parallel extinction.
It represented only a very small percentage of the minerals found.
Zircon occurred as colorless elongated prisms, sometimes terminated
by pyramidal faces*

It nearly always appeared rounded at the crys­

tal edges, particularly the fragmentary grains.

In a few grains

19.
inclusions of a colorless mineral with a lesser index than zircon
were observed.
Tremolite occurred as colorless and white grains.

The habit,

cleavage, structure, and fracture appeared similar to that of
hornblende although its refractive index was lower and the bi­
refringence stronger.

Tremolite was distinguished from diopside

by a lower refractive index.
Muscovite constituted less than 1 percent of the minor min­
erals.

It occurred as thin transparent flakes, marked by a low

bluish-gray interference color yielding a well-centered biaxial
figure.
Tourmaline was present in very small amounts.

The grains

were relatively small colorless prisms sometimes terminated by
pyramids.

Tourmaline was distinguished from other minerals by

its very strong pleochroism, its rather high birefringence, and
its maximum absorption direction.
Presentation of Mineral Count Data
Mineral count da.ta are usually converted to weight per­
centages (3, 12, 16).

The conversion is correct if the grains

within the fraction have the same average volumes and the same
specific gravity.

Coated unknown grains as well as different

mineral shapes would also reduce the accuracy of the conversion.
In these studies the coated unknown grains were few in number.
The shapes of some of the same species of minerals varied tre­
mendously in the C and A horizons.

This is vividly brought out

in the photomicrographs, Plates 3, 4, and 5.

The hornblende is

20.
differently shaped in the A than in the C horizon.

It would seem

therefore, that the conversion of mineral counts of such varying
dimensions to average volumes or weight percentages without cor­
rection would result in great errors.

Mineral counts were not

converted to volume or weight percentages because of the belief
that the specific gravity varies as a mineral weathers and the
observations that the shapes of the minerals and the frequency
of the maximum cross-sectional areas varied within and between
horizons.
To show that the frequency distribution of the maximum
cross-sectional areas is not the same the Extract from Table 6
is given.
_

•
o
H

Extract from Table 6__________________________________________ —
Mn n
i
___________Number of Particles in Various Horizons
and
Maximum cross-sectional area
Horizon
1.0
0.25
0.5
2.0
3.0
4.0
6.0
Dark Green A
Hornblende B

c

96
474
—

5 24
496
1216

Wallac e sand
34
72
88
38
608
246

28
32
19

4
8
10

2
4
13

These results show that in the A horizon the centra.1 tendency
is near the 1.0 maximum cross-sectional area; in the B horizon
between the 0.5 and 1.0; while in the C horizon it is between 1.0
and 2.0.
In the tables and figures to follow, the maximum cross-sec­
tional areas are designated with the class marks; 0.25, 0.5, 1.0,
2.0, 3.0, 4.0, 6.0, and 10.0; and the class numbers 1, 2, 3, 4,
5, 6, 7, and 8.

These class marks and class numbers correspond

21.

to the following maximum cross-sectional areas:
Class No.

Class Mark

Maximum cross-sectional area

1

0 .25

.075)2 /4 mm.

2

0.5

.075)2

3

1.0

.075)2

4

2.0

2 .075)2

5

3.0

3 .075)2

6

4.0

4 .075)2

7

6.0

6 .075)2

8

10.0

10 .075)2

A

Discussion of Results
Garnet
Many workers have studies the use of resistant heavy minerals
as indices of soil development.

In general, when soils weather the

resistant minerals show a relative concentration, whereas the less
resistant decrease in relative abundance or disappear completely.
Zircon, tourmaline, anatase, rutile, magnetite, and garnet are the
most common resistant minerals.
The relative concentration of a resistant mineral can be de­
termined accurately by comparing the size distribution of the en­
tire sample.

This is far too time consuming.

A comparison of a

very narrow grain size may also lead to erroneous conclusions
since the choice of a particular grain size is purely arbitrary.
The latter is brought out more clearly in the Extract from Table 3
and Table 4.

22.

Extract from Table 3 and Table 4
Number of Particles in Various Horizons
Mineral
and
Horizon
0.25

0.5

Maximum cross-sectional area
1.0
2.0
3.0
4.0
6.0

16
6
26

8
2
21

—

Kalkaska sand
74
80
26
14
62
38

30
20
70

13
24
19

2
——

Brown
Garnet

A
B
c

51
10
——

Rubicon sand
26
27
6
2
2
2
3
14
2

Brown
Garnet

A
B
C

39
2
• “*

61
12
24

10.0

6

These mineral counts are of varying magnitude in all the grain
sizes and the choice of any one grain size would not give the com­
plete picture of the apparent garnet accumulation.

In this inves­

tigation the grain size limits were very wide, 0.25-0.02 mm.

It

was believed that a comparison of the grain sizes within this wide
range of grain sizes came nearer to deciding the actual amount of
resistant minerals in podsols than a comparison of a very narrow
grain size range.
In the Roselawn sand, Table 4, the apparent accumulation of
the same size grains in the A horizon occurred in the 1.0, 2.0,
4.0, and 6.C maximum cross-sectional area groups whereas in the
Emmet sand, Table 3, it occurred in the 0.5, 1.0, 2.0, 3.0, 4.0,
and 6.0 groups.
.The garnet grains were more numerous in all horizons of the
Group I than in the Group II soils.
The reduction in the size of the garnet perticles in the A

23.

when compared with the C was greater in the Group II soils.

This

does not necessarily point to a greater degree of weathering in
these soils.

It must he remembered that the Group I soils orig­

inally had a greater number of garnet grains to reduce; therefore
they suffered a greater total breakdown.
There was a greater decrease in size of the garnet grains
from the C to the B, than from the C to the A horizons in all the
soils.

This might indicate a greater severity of weathering in

the B than in the A horizon.
Opaque Minerals
The opaque minerals were a dominant group of the heavy min­
erals in all the soils studied.
hornblende in total quantity.
the opaque minerals.

They rank next to the dark green
Magnetite constituted 90 percent of

Chandler (6) points out that the relative re­

sistance of the more important soil minerals to podsol weathering
is as follows:
resistant
zircon
magnetite
quartz
garnet

moderately resistant
epidote
orthoclase
diopside

easily weathered
hypersthene
hornblende
plagioclase
olivine

Cady (3) similarly found that magnetite and garnet seem to be
little affected by podsolization.
In these studies, Figures 8, 9, 10, and Table 5, magnetite
appeared to weather more readily in the Group I than in the
Group II soils.

A close examination of Figures 8 and 10, and

Table 5 shows that in the Group I soils the mineral grains in
the A horizon have decreased in size and total amount when

24.
compared to the C.

In the Group II soils (except Eastport) the min­

eral grains in the A horizon have accumulated.
The B horizons of the Group I soils in general had greater
amounts of magnetite than the Group II soils.
A comparison of the data for garnet and the opaque minerals,
the two most resistant minerals studied, Tables 3, 4, and 5, and
Figures 2 to 10, indicates that garnet was more resistant than the
opaque minerals.
The presence of the more intense brown B horizon in the Group I
soils has always been difficult to explain.

Data from Table 1 and 2

suggest that this intense brown was due to a greater amount of iron
oxide and organic matter.

Mineralogical examinations, Figures 2 to

16, indicated a greater weathering of the opaque and ferro-magnesian
minerals in the Group I soils.

Birnbaum, Cohen, and Sidhu (l) have

shown that the color change, ranging from yellow to dark brown, of
synthetic iron oxide was caused mostly by particle growth.

It is

believed that the intense brown color due to the inorganic colloids
may be explained by the greater original content of the opaque and
ferro-magnesian minerals s,nd of the greater decomposition of these
minerals in the Group I soils.
Dark Green Hornblende
The mineral count of dark green hornblende is given in Figures
11 and 12, and in Tables 6 and 7.
The A and C horizons of the Group I soils in general contained
much greater quantities than the Group II soils.

The A horizon of

the Emmet sand when compared with the A horizon of the Roselawn

25.

sand had from 2.9 to 13.1 times as many grains.

This comparison is

shown in the Extract from Table 6 and 7.
Extract from Table 6 and 7
Maximum
cross-sectional
area
0.5
1.0
2.0
3.0
4.0
6.0
10.0

Roselawn
A

Emmet
A

20
119
292
22
52
15
2

65
1166
935
63
179
197
21

Emmet A / Roselawn A

3.2
9.8
3.2
2.9
3.4
13.1
10.5

A similar comparison of the Rubicon and Kalkaska A horizons
showed that the Kalkaska A horizon had almost 1-J- times more grains
than the Rubicon.
Eastport sand, because it probably contained recent wind blown
material, did not follow the pattern of the A horizon, Group II
soils.

It contained a very high percentage of small grains; many

more than could have weathered from the C horizon.

Plate 5 illus­

trates some of the small hornblende grains.
The A horizons of Kalkaska and Emmet soils (soils that support
a hs.rdwood cover) showed a greater count and amount of dark green
hornblende than the Wallace, Rubicon, Roselawn, and Grayling soils
(soils that support a pine cover).
With the grain size distribution of the C horizon as a basis
for measuring the grain size change in the B horizon the following
were noted;
(l)

The grains in the B horizons of the Group I soils de­
creased in size to a greater extent than the Group II

26.
soils (except Eastport).
(2)

The grains in the Eastport B horizon decreased the most in
size.

In general it might be stated that the Group I soils have
weathered more than the Group II soils.
Photomicrographs of hornblende, Plate 3, show a greater se­
verity of weathering in the A horizons of Kalkaska and Emmet sands
than in the Rubicon, Roselawn, Grayling, and Eastport sands.
Gray Green Hornblende
The gray green hornblende grains, Figures 13 and 14, and
Tables 6 and 7, were almost as numerous as the dark green horn­
blende.

The Group I soils in general had a greater number of

grains than the Group II soils.

The A horizon of the Eastport

sand, as in the dark green hornblende, contained more grains
than could be accounted for by the weathering of the C horizon.
As mentioned before these additional grains were probably depos­
ited by wind action.
The B horizons showed the unusual trend of accumulation.
There were in most of the soils investigated more grains of all
sizes in the B horizons than there were in the C horizons.

This

might indicate that of all the minerals studied, the gray green
hornblende is the most resistant to weathering in the B horizon.
The A horizons of the Kalkaska and Emmet soils (soils that
support a hardwood cover) showed a greater amount of gray green
hornblende than the A horizons of the Wallace, Rubicon, Roselawn,
and Grayling soils (soils that support a pine cover).

This

0
U r7
I

follows the same pattern as given for the dark green hornblende.
To represent the relative degree of weathering between the
Group X and Group II soils one soil from each group was chosen.
Yfith the grain size distribution of the C horizon as a basis for
measuring the grain size change in the A horizon the following were
noted:
(1)

The A horizon of the Rubicon sand appeared to be more
severely weathered than other soils in Group I.

(2)

The Kalkaska A horizon was the most severely weathered
of the Group II soils*

(3)

The grains in the A horizon of the Kalkaska sand had de­
creased to a greater extent than those in the A horizon
of the Rubicon sand.

Epidote
Epidote was not present in sufficient quantities in any of
the soils except Wallace sand to justify any definite statements
on weathering.

The microscopic count for any one grain size was

usually less than 25 grains, Figures 15 and 16, and Tables 6 and 7.
The A and C horizons of the Wallace sand had a relatively
high content of epidote.

Of all the other soils investigated

Rubicon ranked second in the total count of epidote grains.

The

Extract from Table 6 and 7 illustrates that in the C horizon of
the V/allace sand the epidote grain count was at least three times
as great as the grain count in any of the other sands.

In the

A horizon of the Wallace sand this dominance was considerably
less .

•

28.
Extract from Table 6 and 7
Maximum
crosssectional Wallace Rubic on
C
C
area
0.5
1.0
2.0
3.0
4.0
6.0
10.0

Wallace C
Rubic on

Wallace
A

Rubicon
A

Wallace A
Rubicon A

___

___

400
176
35
16

51
37
13
18

7.8
4.7
2.7
0.9

4
318
30
10
4

121
151
23
6
7

.03
2.1
1.3
1.7
0.6

---

---

---

---

---

---

Minor Minerals
The minor minerals consisted of tremolite, zircon, muscovite,
and tourmaline.

The heavy mineral count of these minor minerals

from 20 grams of organic-free soil was insufficient to provide any
statistically significant comparisons between the Group I and
Group II soils.

The results represented in Tables 8 and 9 did

show that zircon was present in high amounts in the Wallace soil.
This apparent accumulation of zircon might indicate a greater se­
verity of weathering in the B than in the A or C horizons of the
Wallace soil.
c*

NEUBAUER TESTS
Neubauer tests are included to explain a type of weathering
in podsols.

It is believed that the heavy minerals, since they

constitute a much greater variety and amount of mineral species
in these soils, could be utilized to measure the relative growth
of plants.

Recent investigations by Graham (14, 15) suggested

that the organic matter closely surrounding the individual

29.
particles in the B horizons might decompose the fine sand particles
for the nourishment of plants.

Also, it was thought that the pres­

ence of no organic matter as found in the C horizons would provide
little nutrient delivery to plants.
The Neubauer tests as used in these studies provided an indi­
cation of the relative nutrient delivery of the soil.

The procedure

and equipment for conducting the Neubauer tests was that recommended
by Thornton (39).

All cultures were run in triplicate.

Rosen rye

seed from South Manitou, Michigan was provided by the Farm Crops
Section of Michigan State College.
Ceresan.

The seeds were dusted with

The temperature of the air in the Minnesota Seed

Germinator was maintained at 20 (plus or minus one) degrees
Centigrade.

Distilled water was added as required.

After 17

days the rye seedlings were removed, washed, air-dried for two
hours, and weighed.
The results representing the B and C horizons of all the
soils are presented in Plates 1 and 2, and Figure 17.

These

results showed that the growth on the C horizon of all soils
was approximately the same as that on the quartz sand check.
The tests suggest that there was no nutrient delivery from the
soil minerals to the plant.
The growth on the B horizons showed a decided increase
over the quartz sand check.

The greatest growth response was

found on the Kalkaska and Emmet soils, soils that originally
supported a hardwood cover.

This indicated a greater nutrient

delivery from the B than from the C horizons.

30.
Since the sand particles of the B horizons of the Kalkaska and
Eramet soils are higher in coatings of organic matter than most of
the other soils studied, it was believed that the presence of the
organic matter assisted in the decomposition of the heavy minerals.
The following evidence supports the concept that mineral de­
composition in Michigan podsols is due primarily to organic matters
(1)

A greater amount of total heavy minerals in the Group I
than in the Group II soils, Figure 1.

(2)

The higher organic matter content in the B horizons of
the Group I than in the Group II soils, Table 2.

(3)

The greater quantity of calcium and magnesium minerals
in the Group I than in the Group II soils, Figures 11
to 16.

(4)

The greater quantity of dark green hornblende in the C
than in the B horizons of all soils, Figures 11 and 12.

(5)

The greater decomposition of dark green hornblende in
the B than in the A horizon, Figures 11 and 12.

(6)

A similarity of nutrient delivery in the C horizons of
all soils, Figure 17 and Plate 2.

(7)

A variable yet greater amount of nutrient delivery in
the B than in the C horizons of all soils, Figure 17
and Plates 1 and 2.

(8)

A greater organic than inorganic base exchange capac­
ity of the soil; a dominance of the calcium and mag­
nesium cations in the organic exchange complex, the
results of Tedrow and Gillam (38).

Also, a higher

31.

organic exchange capacity of the B horizon of the Group I soils
(represented by Kalkaska and Emmet) than for the Group II soils
(represented by Rubicon).
These data indicate that organic matter greatly assisted the
decomposition of the heavy minerals, particularly the dark green
hornblende in the fine sand fraction.

The magnitude of this de­

composition varied with the amount and type of organic matter.
The presence of relatively high amounts of heavy minerals without
organic matter did not provide sufficient plant nutrient delivery.
SUMMARY AND CONCLUSIONS

A study was made of the heavy minerals in the fine sand frac­
tion of some Michigan podsols.

Detailed data obtained from petro-

graphic observations, mechanical and chemical analyses, and
Neubauer tests were correlated to indicate a type of soil weath­
ering.

A summation of the results and of the conclusions follows:

The brown B horizon of some Michigan podsols was the result
of a vigorous decomposition of a relatively high original content
of the opaque and ferro-magnesian minerals.
Organic matter was an effective weathering agent of some
heavy minerals in the B horizons.
The least resistant mineral to podsol weathering was dark
green hornblende, followed by gray green hornblende, the opaque
minerals, and the garnets.
The relative resistance to weathering varied within the
profile.

In general, the B horizons suffered a greater

32.
decomposition of the heavy minerals than the A or C horizons.
The Kalkaska and Emmet soils (soils that support a hardwood
cover) showed a greater amount of the calcium and magnesium heavy
minerals in all horizons of the profile than the Wallace, Rubicon,
Roselawn, and Grayling soils (soils that support a pine cover).
The Wallace sand showed marked differences in the microscopic
count of heavy minerals from the other soils investigated.
A similar heavy mineral assemblage was found in the Group I
(Wallace, Kalkaska, and Emmet) and Group II (Rubicon, Roselawn,
Grayling, and Eastport) soils, although the quantity of heavy
minerals was lower in the Group II soils.
Quantitative data on the heavy minerals were difficult to
secure.

More accurate results were obtained by subjecting the

entire rather than a small part of the fine sand fraction to heavy
mineral analyses.

33.
BIBLIOGRAPHY

Birnbaum, H., Cohen, H . , and Sidhu, S. S. X-ray diffraction
studies of color variation of iron oxide pigments. Jour. App,
Phys. 18: 27-29. 1947
Buckhannan, W. H. and Ham, W. E. Preliminary investigations
of heavy mineral criteria as an aid in the identification of
certain soils in Oklahoma. Soil Sci. Soc. Amer. Proc. 6:
63-67. 1941
Cady, J. G. Soil analyses significant in forest soils in­
vestigations and methods of determination: 3. Some mineralogical characteristics of podzol and brown podzolic forest soil
profiles. Soil Sci. Soc. Amer. Proc. 5: 35 2-354. 1940
Carroll, Dorothy. Recording the results of heavy mineral
analysis. Jour. Sed. Petr. 8: 3-9. 1938
_________________
Mineralogy of some soils from Denmark,
Western Australia. Soil Sci. 60: 413-426. 1945
Chandler, R. F., Jr. The relation of soil character to
forest growth in the Adirondack region. N.Y. (Cornell) Agr.
Expt. Sta. Ann. Rpt. 54: 93-94. 1941
Chayes, Felix. Petrographic analysis by fragment counting.
Econ. Geol. 39: 484-505. 1944
-___________
A correction factor for specific gravity and
volume differences in fragment analysis. Econ. Geol. 41:
749-761. 1946
Dryden, Lincoln and Dryden, Clarissa. Comparative rates of
weathering of some common heavy minerals. Jour. Sed. Petr.
16: 91-97. 1946
Fairbairn, H. W. Gela/tin coated slides for refractive index
immersion mounts, Amer. Min. 28: 396-397. 1943
Fieger, E. A. and Hammond, J. W. Profile studiesof the
coastal prairie soils of Louisiana: 3. Mineralogical prop­
erties. Soil Sci. Soc. Amer. Proc. 2: 121-131. 1937
Fry, V/. H. Petrographic methods for soils laboratories.
U.S. Dept. Agr. Tech. Bui.
344. 1933
Goldich, S. S. A study in rock weathering.
46: 17-5 8. 193 8

Jour.Geol.

34.
(14)

Graham, E. R. Colloidal organic acids as factors in the
weathering of anorthite. Soil Sci. 52: 291-295. 1941

(15)

Soil development and plant nutrition:
2, Mineralogical and chemical composition of sand and
and silt separates in relation to the growth and chemical
composition of soy beans. Soil Sci. 55: 265-273. 1943

(16)

Haseman, J. F. and Marshall, C. E. The use of heavy minerals
in studies of the origin and development of soils. Mo. Agr.
Exp. Sta, Res. Bui. 3 87. 1945

(17)

Humbert, R. P. and Marshall, C. E. Mineralogical end
chemical studies of soil formation from acid and basic
igneous rocks in Missouri. Mo. Agr. Exp. Sta. Res. Bui.
35 9. 1943

(18)

Jeffries, C. D. A method of preparing soils for petrographic
analysis. Soil Sci. 5 2: 451-454, 1941

(19)

__________
Recent advances in soil mineralogy.
Unpublished. Presented before the meeting of the American
Society of Agronomy and the Soil Science Society of America,
Omaha, Neb., Nov, 21, 1946

(20)

and Y/hite, J. W. Some mineralogical and
chemical characteristics of a Hagerstown soil profile.
Soil Sci. Soc. Amer. Proc. 2: 133-141. 1938

(21)

______________________ . ________ Variations in the
composition of feldspar from a Hagerstown soil profile.
Soil Sci. Soc. Amer. Proc. 3: 26-31. 1939

(22)

____________________________
Mineralogical soil
analysis as an aid in soil classification. Soil Sci.
Soc. Amer. Proc. 4: 364-367. 1940

(23)

___________________ .____________
Some mineralogical
characteristics of limestone soils of different localities.
Soil Sci. Soc. Amer. Proc* 5: 304-308. 1941

(24)

Johannsen, A* Manual of Petrographic Methods. Second
Edition; New York: McGraw-Hill Book Co., Inc. 1918

(25)

Krumbein, W. C. and Pettijohn, F. T. Manual of Sedimentary
Petrology. New York: Appleton Century. 1938

35.

(26)

Krumbein, W. C. and Rasmussen, W. C. The probable error of
sampling beach sand for heavy mineral analysis. Jour. Sed.
Petr. 11s 10-20. 1941

(27)

Marshall, C* E. A petrographic method for the study of soil
formation processes. Soil Sci. Soc. Amer. Proc. 5; 100-103.
1941

(28)

and Jeffries, C. D. Mineralogical methods
in soil research: 1. The correlation of soil types and parent
materials with supplementary information on weathering
processes. Soil Sci. Soc. Amer. Proc. 10: 397-406. 1946

(29)

McCaughey, W. J. and Fry, W. H. Microscopic determination
of soil forming minerals. U.S. Dept. Agr. Bur. Soils Bui.
91: 5-100. 1913

(30)

Mickelson, G. A. Mineralogical composition of three soil
types in Ohio with special reference to changes due to
weathering as indicated by resistant heavy minerals. Ph.D.
Thesis, Ohio State University, 1942. Abstracts of Doctoral
Dissertations No. 40* Ohio State University Press. 1943

(31)

Milner, H. B. Sedimentary Petrography.
New York: Nordeman Publishing Co. 1940

(32)

Pye* Willard D. Rapid methods of making sedimentational
analyses of arenaceous sediments. Jour. Sed. Fetr. 13:
85-104. 1943

(33)

Richard, J. A. and Chandler, R. F., Jr. Some physical and
chemical properties of mature podzol profiles. Soil Sci.
Soc. Amer, Proc. 8: 379-3 83. 1944

(34)

Rittenhouse, G. Curves for determining probable errors in
heavy mineral studies. Nat. Res. Council Comm, on Sedimentation
Kept. Exhibit F: 97-101. 1939-1940

(35)

Robinson, W. 0.
7-11. 1945

(36)

Spurway, C. H. Soil testing; a practical system of soil
fertility diagnosis. Mich. Agr. Expt. Sta. Tech. Bui. 132
(2nd Revision). 193 8

(37)

Swineford, Ada and Swineford, Frances. A comparison of
three sieve shakers. Jour. Sed, Petr. 16: 3-13. 1946

Third Edition;

The fusion analysis of soils.

Soil Sci. 5 9:

36.
(3 8)

Tedrow, J. C. F. and C-illam, V/. S. The base-exchange
capacity of the organic and inorganic fractions of several
podzolic soil profiles. Soil Sci. 51: 223-233. 1941

(39)

Thornton, S. F. Soil and fertilizer studies by means of
the Neubauer method. Ind. Agr. Expt. Sta. Bui. 399. 1935

(40)

Truog, E., Taylor, J. R., Jr., Pearson, R. W. , Weeks, M. E.,
and Simonson, R. W. Procedure for special type of mechanical
and mineralogical soil analysis. Soil Sci. Soc. Amer. Proc.
1: 101-112. 1937

(41)

Veatch, J. 0. Agricultural land classification and land
types of Michigan. Mich. Agr. Expt. Sta. Spec. Bui. 231
(1st Revision). 1941

Table 1— Content of Free Alumina, Iron Oxide, and Colloidal Silica in
the B Horizons.

Percent in Organic Free Soil
Soil
Alumina

Iron Oxide

Colloidal Silica

GROUP I
Wallace sand

0.72

0.32

0.10

Kalkaska sand

0.30

0.33

0.04

l&zunet sand

0.07

0.31

0.15

Rubicon sand

0.68

0.25

0.20

Roselawn sand

0.36

0.30

0.12

Grayling sand

0.57

0.30

0.11

Eastport sand

0.09

0.10

0.03

GROUP II

Table 2— Mechanical Analyses, Content of Organic Matter, and pH of
Some Michigan Podsol Soils,

Soil Type
and
Horizon

Percent Perc ent of Separates in Organic Free Soil
pH
of
Organic
Sand
Silt
Clay
Matter (2.0-0.02mm)(0.02-0.002mm)(l ess than 0.002mm)
GROUP I

Wallac e
sand

A
B
C

2,0
0,8

Kalkaska
sand

A
B
C

2.8
0.8

Emmet
sand

A
B
C

2.5
0.4

--

---

96.2
96.4
98.4

2.3
0.6
0.8

1.0
0.9

6.5
5.5
6.5

96.5
94.9
98.1

1.7
2.3
1.4

1.4
1.0

6.0
4.8
6.0

95.7
96.0
98.3

2.0
1.7
1.2

1.2
1.0
---

6.3
6.2
8.0

GROUP II
Rubicon
sand

A
B
C

1,7
0.1

96.4
97.1
100.1

1.0
0.5
0.5

0.6
0.1
---

7.0
6.0
7.0

Roselawn
send

A
B
C

1.8
0.1

98.1
97.1
99.7

1.1
0.7
0.6

0.7
0.5

6.6
5.7
6.6

Grayling
sand

A
B
C

1.9
0.2

98.5
97.5
99.7

1.0
0.6
0.6

0.7
0.7

A.
B
C

1.1
0.1

99.6
99.4
100.1

0.9
0,3
0.3

0.1

Eastport
sand

---

-------

6*3
5.3
6.3
6.2
6.8
6.4

Table 3— Heavy Mineral Count of Brown, Pink, and Colorless Garnet in
the Group I Soils.

Humber of Particles in Various Horizons

Mineral
and
Horizon
0.25

0.5

Maximum cross-sectional area
1.0
2.0
3.0
4.0

6.0

10.0

Wallace sand
Brown
Garnet
Pink
Garnet
Golorless
Garnet

A
B
C

—

A
B
C
A
B
C

—
—
—
—
——
——

82
22
92

22
6
32

26
8
32

14
6
19

—

6

154
12
109

12
—
—
20
——

74
12
16
168
16
70

86
8
77
104
18
22

16
6
38
18
2
6

4
4
19
18
2
10

4
—
12
6
6
19

—
—
—
10
10
22

80
14
38
22
56
14
6
2
14

30
20
70
19
22
14

13
24
19
19
16
5

—
2
—
--

2
24

2
2

13
12
6
15
8
3
2
4
6

34
4
19
11
10
6
4

15

2
22

—

Kalkaska sand
Brown
Garnet
Pink
Garnet
Colorless
Garnet

A
B
C
A
B
C
A
B
C

-—
—
—

39
2
—
—

—

—
37

»
-

mm

7

61
12
24
39
12

74
26
62
26
36
48

24
4
67

4
2
65

_ _

5
_ _

2

Emmet sand
Brown
Garnet
Pink
Garnet
Colorless
Garnet

A
B
C
A
B
C
A
B
C

-

-

-

4

_ _

-

_ —

__

38
8
—
55
2
—

—

6

23
2

59
12
—

84
6
3
23
10
10

—

14

——

14
17
12
10
4
2
17

3
—
—
—

-—

--

'

40.
Table 4 — Heavy Mineral Count of Brown, Pink, and Colorless Garnet in
the Group II Soils.

Mineral
and
Horizon

Brown
Garnet
Pink
Garnet
Colorless
Garnet

Brown
Garnet
Pink
Garnet
Colorless
Garnet

Brown
Garnet
Pink
Garnet
Colorless
Garnet

Brown
Garnet
Pink
Garnet
Colorless
Garnet

Number of Particles in Various Horizons

A
B
U
A
B
C
A.
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A.
B
C
A
B
C
A
B
C
A
B
C

0.25

0.5

— —

51
10
—
PC

— —

——
—

m
mmm

_

M *
■
*

_
_

-

-

__

_

Maximum cross -sectional area
2.0
1.0
3.0
4.0
6.0
Rubicon sand
26
27
2
2
14
3
30
45
2
2
11
3
5
15
2
2
3
3
Roselawn sand
10
30
5
1

-

1
7

14
10
6
10
34

1

5

8

-

-

-

— -

-

_ _

15

Grayling sand
20
45

1
26

3
10

_
_

—

mm

_ _

__

—

mm

-

7

2

7
8
2

8

-

65

Eastport sand
18
22
~ —

-

_

_

13

« i
n

■e*

---

---

_ _

4

_

-

7
1

42

16
6
26
11

8
2
21
2

26
1

11
_
_
_
_

6
——
—
— mm
mm mm

—

4

56
22
8
4
12
6
2
4
3

29
4
19
10
8
6
7

11
2
5
4
2
5
5

6

1

6
2
7
8
4
5
5
2

2
4
17
8
O
21

1
OAf
<
9
8
2
13

3
2
11

7
2
59

2
2

4
10
—

10
— —

2
6

2
—
2

2

-

3
_

18

_

—
—

_ —

4
3

-

24
1

— —

2

6
2
2
8
4
6
1
2

10.0

31

53

20

9
—

o
7

2

7

—

- -

—

--

—

Table 5 — Heavy Mineral Count of the Opaque Minerals.

Humber of Particles in Various Horizons

Soil Type
and
Horizon
0.25

0.5

Maximum cross- sectional area
3.0
4.0
1.0
2.0

6.0

10.0

2

2

4
36
7
8
30
39

28
24
11
24
28

4
26
24
7
12
8

_
__
—
5
2
6

—

—

10
25
10
16
15

8
7
7
24
—

GROUP I SOILS
Wallac e
sand
Kalkaska
sand
Emmet
sand

A
B
C
A
B
C
A
B
C

_

_

__
—

46
60
74
70

—
—
—

—

3
26
—

890
150
5 89
398
362
232
567
112
446

172
108
624
178
406
475
399
200
143

18
36
240
19
17 8
103
147
40
666

22
20
29
7
13 8
50
19
52
110

7
18
27
40
58
8
35
26
26
39
16
13

20
22
47
-28
78
11
15
60
95
1
20
26

GROUP II SOILS
Rubic on
sand
Roselawn
sand
Grayling
sand
Eastport
sand

A
B
C
A
R
C
A
B
C
A
B
C

__
__
__
___
__
—

__
_
—

430
82
7
60
25
4

368
10
8
116
54
75
83
24

_ _

—

39

13

_

_

22

271

60
32
53
275
118
50
102
60
86
25
2
341

42
Table 6— Heavy Mineral Count of Dark Green Hornblende, Gray Green
Hornblende, and Epidote in the Group I Soils.

Number of Particles in Various Horizons

Mineral
and
Horizon
0.25

0.5

Maximum cros s-sectional jar ea
6.0
1.0
2.0
3.0
4.0

10.0

Wa1lac e san d
Dark Green
Hornblende
Gray Green
Hornblende
Epidote

A
B
C
A
B
C
A
B
C

96
474

—

_

_

_

_

4
230

—
4
22
—

—

—
—

524
496
1216
150
25 2
77
318
18
400

72
88
608
12
114
10
30
18
176

34
38
246
18
60
32
10
8
35

28
32
19
12
66
16
4
2
16

24
160
86
19
66
34
19
28
14

17
64
202
42
14
11
10
14

12
5
6
6
5

19
28
31
15
18
19
6
2
2

21
10
193

4
8
10
10
30
16

2
4
13
2
4
6

—

2
—

—
—

Kalka ska sand
Dark Green
Hornblende
Gray Green
Hornblende

A 185
B
C
A 1C 18
B
C
A
B
C
—

—

—

_ _

—

Epidote

—

16 84
56
96
2405
32
12
37
2
5

120
264
1138
26
464
298
13
28
43

50
208
1238
20
102
120
15
36
17

M

19
58
53

Ml

Emmet sand
65
68

Dark Green
Hornblende

A
B
C

Gray Green
Hornblende

A
B
C

w

Epidote

A
B
C

- -

—

—

—

_ _

_
- -

—

53
46

1166
198
83

935
136
13 8

63
56
33

179
32
396

197
26
536

1109
300

113
178
110

4
192
110

53
106
28

29
24
17

69
20
3

2
16
3

27
10
28

—

—

4

—

65
10
- -

W

10
19

11

—

—

—

17

6

43.
Table 7 --Heavy Mineral Count of Dark Green Hornblende, Gray Green
Hornblende, and Epidote in the Group II Soils*

Number of Particles in Various Horizons

Mineral
and
Horizon

Dark Green
Hornblende
Gray Green
Hornblende
Epidote

Dark Green
Hornblende
Gray Green
Hornblende
Epidote

Dark Green
Hornblende
Gray Green
Hornblende
■■
%
Epidote

Dark Green
Hornblende
Gray Green
Hornblende
Epidote

0.25

A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C

—
—
—
—
—
—
__
__

0.5

cross--sectional area
Maximum 1
3.0
4.0
1.0
2.0

455
15 2
5
281
330
11
122
—
—

Rubic on sand
141
131
14
22
59
105
37
16
10
16
59
36
151
23
—
-11
21

20
200

—
——
—
—

__

—

7
240
25
4
—
3
68
14
3
10
8

—
—

—

—

—

130

783
2
48
2769
2
7
7
—

—

7 80
__
—
—
—

—

Roselawn
119
102
200
53
46
64
51
2
25

sand
292
110
269
26
78
3
30
12
38

Grayling sand
230
95
100
314
3
34
80
32
48
34
8
16
115
59
—
—
13
Eastport sand
364
131
8
251
35 9
46
546
10
40
59
1C
3
—
—
37
51

6.0

1C. C

14
32
140
2

5
20
23

15
20
5
4
12
5
6
-6

29
28
161
7
10
57
7
26

11
—
4
27

22
46
40
4
130
4
10
4

52
96
90
2
100
1
9
8
1

15
12
20
2
6
13
5
4
3

6
5
—
9

10
16
18
9
30
3
20
6
18

30
18
107
7
60
47
11
4
23

8
56
88
8
10
23
7
2
13

15
3
26
3
14
4
4
2
8

55
16
42
4
18

39
10
42
4
34

8
26
22

26
46
—
14
34
—
—
—

O

5
2
13

—

—

18
--

11
4
18

—

8

p.
11
—
—
—
2
6
1

—

44.
Table 8--Heavy Mineral Count of the Minor Minerals:
Tremolite,
Zircon, Muscovite, and Tourmaline in the Croup I Soils.

Mineral
and
Horizon

Number of Particles in Various Horizons

0.25

Maximum cross-sectional area
1.0
2.0
3.0
4.0
6.0

0.5

10.0

Wallace sand
Tremolite

Zircon

Muscovite

Tourmaline

A
B
C
A
B
C
A
B
C
A
B
C

—

-

34

—
—
—
—

290
3
—

6
20
150
22
30

—
—

3
—
50
3

mmw

2
2
10
34
10
10

2

—

2

—
12
—.

3
2
4
3

2
8
3

32
6
10

—
10
—

—
mmmm

mmmm

—

6
e
—
15
445
—
—

4
—
6
36
5
—
—

2
—
—
4
22
2
—
—

—

—

mmmm.

Kalkaska sand
Tremolite
Zircon

Muscovite
Tourmaline

A
B
C
A
B
C
A
B
C
A
B
C

__
-

9
8

—
3C

9
12
—
33
-—
19
6
2

6
46
13
24
—
7
2
8
5

—
----

2
12
0
6
84
2
—
—
7
2
—
o
fO

2

—
2
—
10
—
—
—

2
—

—

Emmet sand
Tremolite

Zircon

Muscovite
Tourmaline

A
B
C
A
B
C
A
B
C
A
B
C

-

—
~—

11
2

4

6
20

--------

—

4

------

—

—
—

--- -

-

13
—

- -

6
6

—

2.
- -

14
—
6
-6

14
3
4
--

—

—

—

—
»

mm.

—
—
3
6

10
—

—

—

—

6

p

——

-

—
"***
—

—
-—
—
—

8
2

6
—
--

6
6

—

—

45.
Table 9— Heavy Mineral Count of the Minor Minerals:
Tremolite,
Zircon, Muscovite, and Tourmaline in the Group II Soils,

Mumber of Particles in Various Horizons

Mineral
and
Horizon

0.25

0.5

Maximum cross -sectional area
1.0
2.0
3.0
4.0
6.0

10.0

Rubicon sand
A
Tremolite B
C
A
Zircon
B
C
A
Muscovite B
C
A
Tourmaline B
C

—

2

~

—

2
2

1
30

5

—
—

--

2
2
2
10
2

9

—

—

—

—

--

2
2

---

6
3
6

—

—
—

4

3

—

—

10

—

2
—

—

6
1

2

2
—

8
1
6

2
—

2

4
—

—

—

—

—

—

—

—

—

Roselawn send
Tremolite

A
B

Muscovite

O

8

—

C

Zircon

3

1

—

——

2

—

6

—

3

A
B

—

C

—

A
£

—

2

—

C

—

A
Tourmaline B

—

C

—

—

12
6

10
8

3

3

9

6
-1

_
—

10

2
4
1
2

3

2
—

3

—

—

—

3

—

—

—

—

—

—

—

—

—

—

—
—

3

2

—

Grayling sand
Tremolite

Zircon

Muscovite

A
Jfl
B
C
A
B
C
A
B

4
—

--

—

2
2

—
—
—
—
—

C

—

A
Tourmaline B
C

—

—
—

—

1
—

1
2

1
—
—

1

—

2

2
3

—

—

—

—

—

—

—-

—

—

—

2
3
--

—

—

1
—

—

—

—

3

Table 9 (Cent.)

A
Tremolite
B
____________ G
A
Zircon
B
____________ C
A
Muscovite
B

—
—
—
—

Eastport sand
1
3
4
-—
—
9
4
7
8
8
—
—
—
4
2
52
20
20

_____________ C

Tourmaline

A
B
C

2

—

20

1

—

—

9

2

1
—

-—

5
—

—

—

13

—

4
~
6

-—

~

47.

Plate 1.

TsTeubauer tests on the 3 horizon of some Michigan podsols.
U p p e r - — 25 Kalkaska
Middle— 26 Grayling
Lower
23 Eastport

31 Emmet
29 Rubicon
35 Check

21 Wallace
18 Roselawn

PI f-te 2.

Neubauer tests on the G horizon of some Michigan podsols.

1 Kalkaska
2 Emmet
3
-- Wallac e
4
-- Chec k

5
6

Eastport
Roselawn
7
-- Rubic on
8 Grayling

49.

Plate 3.

Photomicrographs of hornblende, 70X
U p p e r - --severely weathered hornblende in the A
horizon of the Kalkaska and Emmet sands.
Lower
slightly weathered hornblende in the A
horizon of the Rubicon, Roselawn, Grayling,
and Eastport sands.

*±.

Photomicrographs of the heavy minerals of the C horizons,

VOX

Upper---slightly altered heavy minerals of the Group I soils.
Lower--- slightly altered heavy minerals of the Group II soils.

m

If

Plate 5.

Photomicrographs of "the heavy minerals. 7CX
Upper
-veil-rounded grains characteristic of all
horizons of the Wallace sand.
Lever
small, irregular, wind-blown hornblende grain
in the A horizon of the Eastport sand.

•3Q

Percent

of Heavy Minerals

.25

CBA
Wallace

Figure 1-

CBA

CBA

CBA

Kalkaska

Emmet

Rubicon

Percent of

heavy

C BA

C BA

Roselawn Grayling

CBA
Eastport

minerals
in the fine send fractionof the
Group I and II soils.
(2C grams of organic-free soil?

53

-= f C\J

f\J OJ C\J C\J Oj

OJ (\1

I
COI i— r— r~ o— r-r— r— i—
<D|
Uj o o o o o o o o
<
•
■
0i
I f M i M f M f M i

I—I OJ

vX) c

garnet

f\(

\ IT\ LT\ IT>

lT \v _ 0 r— CO

icroscopic count end grain size distribution of brown
in the fine Jind fraction of the Group I soils.
(20 grains of organic-free soil)

o
D

•H

Pm

seiopqjud jo aequm^

•h c\j to 'f »r> <o

co

o
o

oa

o

o

fl0to]!)j9j jo jequmfi

Mlcroacoplc coant and grata alia distribution of pink garnet
in the fine aand fraction of the Oroop I solla.
(20 grant of organic-free soil)

'

figure 3.

o* o • o • o • o • o o o
oj rj

Max. cross-sect
•A « > « M A

Figure 1.

Class No.
CMCMCMCMCMCMCMCM

o o o o o o o o

Cs-S P* c**t— c*- c—
l/M A

OJ

I
I I
I
t I
I I
I •
I I
II
II
Microscopic count and grain size distribution of colorless
garnet in the fine sand fraction of the Group I soils.
(20 grams of organic-free soil)

- i t CM

jo aaquiriN

-^

O
nO

HCVJC'N-^tfXvOP-OO

CO

o

st)To T^Jtb^ jo uoquirvM
soilj

O

of organic-free

CM

O

grams

O

(20

UMITN »T\tf\U"\*f\CT\UM
C^C'- o- c- o c*- cO

Microscopic count and grain size distribution of brown garnet
in the fine sand fraction of the Group II soils.

Max. cross-sect.
O O

Figure 5*

plass No.

CM

CM CM CM CM CM CM CM CM

O
O

o
•>
M

9Q
o

Microscopic count and grain size distribution of pink garnet
in the fine sand fraction of the Group II soils.
(20 grams of organic-free soil)

i

Figure 6.

57.

OJ

Cf\ir\UMT\»rMJNUMf\

CM (N OJ C\< <M CJ

o- S o* c— c— c*~o- o-

o o o o o o o o
OJ « ^ 4 m > 0

8 ©-[0 -Fq.JT9d

JO

aoqtonN

(V CO -J- sO O

X
3
a

o

I
I I
I I
I I
I

«

cn
m
ci

(— I CM CO

O

IPs sO C" CO

r

f
H

O

ci
t~.

cs

Figure 7.

_

t:
j
c

Microscopic count and grain size distribution of colorless
garnet in the fine sand fraction of the Group II soils.
(20 grams of organic-free soil)

OJ
■p
o
a>
<n
OIOJOJOJC^OJOIOJ
1
CO
(0 td U M A tfN l f \ WPS IPS IPS IPS
o 0) c— f-r'-c^c^c^-c'-c't->
o o o o o o o o
o <d

o

CO

~c>

vO

o<f
eexo^^jtBci jo aoqum^

600

Class No

cross-sect.
area

50Q,

400

30Q

Number

of Particles

—

(.075) / 4 mm
(.075}2 / 2
(.075)2
2(.075)2
3(.075)2
4(.075)2
6(.075)2
10(.075)2

200

100.

12345676
W&llaee

Figure P.

Kalkaska

Emmet

Rubioon

Roselawn

Crayling Eastport

Microscopic count and grain size distribution of the
opaque minerals in the fine Band fraction of the
A horison of the Croup I and II soils.
(20 grams of organic-free soil)

60Q

Number of Particles

50Q

Claes No

Max. cross-sect,
area
(.075) / 4
(.075)2 / 2
(.075)2
2(.075)2
3(.075)2
4(.075)2
6(.075)2
10(.075)

123 456 7 8
Wallace

Figure 9.

Kalkaska

Emmet

Rubicon

Roselawn

Grayling

Eastport

Microscopic count and grain size distribution of the
opaque minerals in the fine sand fraction of the
B horizon of the Group I and II soils.
(20 grams of organic-free soil)

60Q

Class No.

(.075)2

(.075)2
2(.075)2
3(.075)2
4(.075)2
6(.075)2
10(.075)2

400

300

Number

of Particles

500

Max. cross-sect.
area
(.075 )'

200

10Q

1234567d
Wallace

Kalkaska

Figure 10.

Emmet

Rubicon

Roselawn Grayling

Eastport

Microscopic count and grain size distribution of the
opaque minerals in the fine sand fraction of the
C horizon of the Group I and II soils.
(20 grams of organic-free soil)

.

p
t*
a>
to
u>
V)

trt
a>
tto nJ

o

0
»v

.V tNJ tv tV vV kV kV tV
U N

•

*
3

U N

U"N U N U N U N

U N

O

o o

U N

c— t'- r-r~ c—C'-t'-c~-

tN o

*

O

•

1~> o

♦

•

•

•

»

^fV4^0 0

•St

d

to

to

0

>r
H

4

<U
(1 o

<<.
N
t CO - -UNv£) f'- oo

«J
o

t>0 (/.

J
—I

t

At 1
—
1
d
aJ A.
XJ d
o
V-i d
o o

-p
4)

to

1'

d a> rH
o _c •H
•H -p o
to
V.
a>
,0 o
a>
•H
d d
d
P o C-t
to *H
t
•H -P o
X) u •P
aJ
a) d
3
N —
i
bO
d
•H
X) o

(0

d
•H

s
(0

d

0)

<p

o

co

d §
«p d
s <u bO
-C o
tv
bO

to
CO

xJ

T

T -t

-p -p

o

o

CO ci

0

>
.

a

Q>
d
cx
oi
—i
o x>
to d
o d
d o
o

o XJ
*H
a>

♦H
=3

s*
to

a>
d

•SH)
t
t
,

<ti

o

CO ci

tU)

t\i

to

U3

■0+
;o

3ax °Tr3-*1xid jo ■Jo'jiinH

d

•H

JC

geXQT^^^d J° Joqmnj{

Class No.

Figure 12.

Microscopic count and grain size distribution of dark green
hornblende in the fine sand fraction of Group II soils.
(20 grams of organic-free soil)

Max. cross-sect
area

-tv O

H c\ r\ ^}-iA>o c^oo

tO
B 9T O ^ J « , {

Microscopic count and grain size distribution of gray green
hornblende in the fine sand fraction of the Group I soils.
(20 grams of organic-free soil)

CM

Figure 13-

6-»

CVCMCv JCMCVCNJ Oi CM

O

o

JO

CM
JdURlN

Figure 11.

liicroscopic
hornblende

count and grain size distribution of gray green
in the fine sand fraction of the Group II soils.

65,

I
In
HI
O

c_>

2*
-j

vM

(tJ
U>
U
it>
I A

iV C-A

o

I I

<d
O-4
H^
1A
O- rO

r

O
CXJI

Microscopic count and grain size distribution of epidote
in the fine sand fraction of the Group I soils.
(20 grams of organic-free soil)

4->
o
0)
in

Figure 15.

66,

:
H
i

iv

^

o o
• •
U ~\
o -

.

><
CO

vO O

I I

Ol*I
v
jo aaquinH

Figure 16.

Microscopic count and grain size distribution of epidote
in the fine sand fraction of the Group II soils.
(20 grams of organic-free soil)

67,

s

ot rj-J-uj jo

j

etiumfl

Figure 17.

rr’ ni

I

luwxj J® ouiMay
Neubauer tests.

Increase of plant material over check.

« pi