DISTéNCTIVE PERSONAUTY AURIBUTES '0?
CHILDABUSING MGTHERS

Thesés for the. Degree of M. A.
MiCHlGAN STATE UNIVERSQTY
BARRY MELNICK
1968

 

m 7 "3

HE:

 

  

.n
33*!

62/215

NB :2 ‘8 1996

200

 

 

 

 

ABSTRACT
DISTINCTIVE PERSONALITY ATTRIBUTES OF CHILD-ABUSING MOTHERS

by Barry Melnick

Ten mothers suspected of severely beating their
children, abusive mothers (AM's), and ten control mothers
(CM's) matched for age, social class, and education were
compared on selected psychological measures to identify
potentially differentiating personality attributes. The
particular psychological measures were chosen because they
were best suited to test certain hypotheses offered in the
literature concerning child abuse. The child-abusing
mothers were obtained through The Wayne County Clinic for
Child Study. All the mothers participating in the study
were largely from the lowest socioeconomic level. Eighty
per cent of the AM's and 100 per cent of the CM's were Negro.

The TAT was administered to measure six psychological
needs, five areas of frustration, and also to obtain a
general measure of pathogenicity based on both the mothers'
capacity to administer to the needs of others and their per-
ception of the environment's willingness to gratify their
own needs. The Family Concept Inventory (FCI) and The
Manifest Rejection Scale (MR) were used to Obtain information

concerning both their perception of the family situation
and their attitudes toward children, respectively. Five

Barry Melnick
California Test of Personality (CTP) subscales were also
employed and included measures of self-reliance, self-
esteem, sense of personal freedom, feelings of belonging,
and withdrawing tendencies.

Six of the 19 measures statistically differentiated
between the AM's and the CM's. These measures included
three of the 12 TAT variables, one of the five CTP measures,
the FCI and the MR indices. ~Analysis of the intercorrelational
findings among the six variables which statistically
differentiated between the two groups revealed that the FCI
intercorrelated with more of the other measures than any
other measuring instrument and accounted for the largest
portion of the total variance. The TAT pathogenic index
had the second largest number of significant intercorrela-
tions and accounted for the next greatest portion of the
total variance. The CTP Self-Esteem and TAT dependency
frustration variables were the next most salient measures.

The high pathogenic scores of the AM's suggests
that they have difficulty empathizing with or administering
to their children's needs. Also, the high pathogenic index
of the AMb indicates that they tend to perceive their own
emotional needs as being frustrated by the environment. The
AM's expressed greater frustration of their dependency needs
in their TAT stories. This finding suggests that they have

strong unmet dependency needs. It was speculated that the
intensity of the AM's unmet dependency needs causes them to

be narcissistic and consequently incapable of identifying

Barry Melnick
with their children. Also, the AM's revealed low self-
esteem on the CTP. The high intercorrelation between the
CTP Self-Esteem subscale and the pathogenic index suggests
that there is a relationship between self acceptance and
the capacity to accept others. The FCI data revealed that
the AM's feel that their present family situation is not
sufficiently meeting their needs. The fact that the PCT
accounted for the largest portion of the total variance
reflects the importance of the present family situation in
understanding child abuse.

An uneXpected finding was that the AM's eXpressed
less rejecting attitudes toward children on the MR than the
CM's. Two possible explanations for this finding are that
the AM's may have been defensive when responding to test
questions for fear that they might be prosecuted or that
they may have had rigid defenses against the expression of
negative feelings.

The results did not confirm the following hypotheses
concerning the dynamics of child abuse: The AM's are
(a) chronically hostile and aggressive individuals
(b) highly frustrated because of their feeling of being
restricted by maternal responsibilities and (c) authoritative
and domineering individuals who engage in a continuous power
struggle with the child.

Because of the use of mainly lower class Negro gs,

the results of the study may have limited generalization.

Barry Melnick
Therefore, the study should be repeated using a larger

sample with a wider representation of social class.

MA.Thesis Committee: Lucy C. Ferguson, John R. Hurley
(Chairman), and Arthur A. Seagull.
Date Accepted: February 2, 1968

 

DISTINCTIVE PERSONALITY ATTRIBUTES OF CHILD-ABUSING MOTHERS

By
Barry Melnick

A THESIS

Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of
MASTER OF ARTS
Department of Psychology

1968

CL+SSWR

ACKNOWLE DGEM ENT S

The invaluable assistance of Dr. John R. Hurley
throughout the entire course of this study is deeply and
sincerely appreciated. I also wish to thank Dr. Lucy C.
Ferguson and Dr. Arthur A. Seagull for reading and

criticizing this manuscript.

ii

List of Tables . .
List of Figures .
List of Appendices
Introduction . . .
Method . . . . . .
Results . . . . .
Discussion . . . .
References . . . .

Appendices . . . .

TABLE

OF CONTENTS

iii

Page
iv

vi

10
22
28
34

38

LIST OF TABLES

Table Page
1. Data on both abusive and control groups . . . . 12

2. Personality measure comparisons of the
abusive and control mothers . . . . . . . . . 23

3. Product-moment correlations anong
statistically significant measures . . . . . 25

iv

LIST OF FIGURES

Figure Page

1. Typal Analysis of personality measures
which discriminated abusive from control
methers O C O O O O O O O O O O O O O O O O O O 27

LIST OF APPENDICES

Appendix Page

A. The Seven Income Categories Used to
Derive Social Class . . . . . . . . . . . . . . 38

B. Raw Scores for Individual §s on CTP, FCI,
MR, and Pathogenic Ratings . . . . . . . . . . 39

C. Comparison of Psychological Need
Percentages Derived from the TAT . . . . . . . 40

D. Comparison of Need Frustration Percentages
Derived from the TAT . . . . . . . . . . . . . 41

vi

INTRODUCTION

Recently, there has been growing interest in the
psychological and social factors involved in child abuse.
The growing interest has encouraged increased investigation
into the underlying psychological and social influences on
the problem. Because the vast majority of these studies
have employed neither quantifiable measures nor control
groups, their results are somewhat questionable.

It has been frequently reported in the literature
that the parents of abusive parents were neglecting, reject-
ing, or sadistically cruel (Komisaruk, 19667 Lazarus, 1966;
Nurse, 1964; Steele, Pollock, & Davoren, 1966; Young, 1953).
Komisaruk (1966) studied a predominantly lower socioeconomic
class population through psychological tests and psychiatric
interviews and reportsthat frequently there was an actual
or psychological loss of a significant figure within the
first five years of life. Also, at least one of the parents
was often sadistic. [Nurse (1964), using social histories,
compared the childhood experiences of the abusive parents
with those of their spouses and discovered that the vast
majority of abusive parents suffered from some form of
emotional deprivation. Steele, 55131.,(1966), using
psychiatric interviews, psychological testing, and home

observation, made an intensive study of 50 families in

2

which child abuse occurred. He found that the parents of
the abusive parent disregarded the needs of their children
and that the abusive parent was forced to satisfy the needs
of his own parents without being able to actually suffi-
ciently satisfy them. DeFrancis (1963) hypothesized
//%Parents who abuse their children are parents who react
I violently to their own unmet dependency needs (p.3).” In a
study of 29 families through observation and interviews, it
was discovered that the abusive parent still had intense
longings for maternal nurturance (Marian & Gould, 1963).

Curtis (1963) suggested that insufficient love and
[aggression toward children were actually transferred from
generation to generation. He felt that children who were
treated with violence which couldn't always be anticipated

learned to react with blind violence to all stress.
There has been considerable theoretical speculation

and empirical research into the psychological factors
underlying child abuse. A few investigators have emphasized
the importance of ego insufficiency in understanding a
parent's aggressive outbursts toward his children. Mental
illness and mental retardation have been posited to help
explain the parent's inadequate judgment and poor impulse
control (Nathan, 1965). However, the more extensive
psychiatric studies have failed to find a significant degree
of psychosis or'mental retardation (Komisaruk, 1966; Steele,

gt. _a_l__. . 1966).

3

Several studies have found the abusive parent to be
extremely immature and characterized by impulsivity,
narcissism, and strong dependency needs (DeFrancis, 1963:
Kempe, Silverman, Steele, Droegemueller, & Silver, 19627
Komisaruk, 19667 Lazarus, 1966). The abusive parent was
conceived as an emotionally unstable person with ungratified
needs who found children a burden. The parent's narcissism

prevented the establishment of any real attachment between
parent and child. Thus, the abusive parent is thought to

leek the capacity to tolerate the natural frustrations aris-
ing from child rearing.

The concept of role reversal has been employed to _.
dynamically explain the act of child abuse (martin & Gould,
1963; Steele, gt 31., 1966). Steele, g§_a;.,(l966) asserted
that personality traits, such as immaturity, impulsivity,
and dependency were too pervasive in the general p0pulation
to adequately explain the cause and effect relationship of
child abuse. 'It is felt that the abusive parent invests
his child with the same reactions and attitudes which were
once held by his own parents toward him during his childhood.
‘The abusive parent was never able to please or satisfy his
parents and felt continually criticized and persecuted by
them. Furthermore, the abusive parent's own parents dis-
regarded the needs of their children. In the abusive
parent's present family situation, Steele, §§,gl., (1966),
using psychiatric interviews, psychological tests, and home
observation on 50 families, found that the abusive parent

expected the child to satisfy his unmet emotional needs in

 

 

4

order to make up for the deprivation he suffered as a child.
The abusive parent eXpected the child to comply with his
needs and looked to the child for approval to an exaggerated
degree. *When the child cried despite efforts to comfort
him or was difficult to handle, the abusive parent mis-
interpreted this behavior as criticism, willfulness, or un-
responsiveness which resembled his own parents' reactions
toward him. This resemblance reawakened the hostile feel-
ings he had had for his own parents. Thus, the concept of
role reversal involves the ideas that the abusive parent
has strong unmet dependency needs and that the act of
battering the child is displacement of hostility. '

It has also been posited that rather than being
merely impulsive and immature, the abusive parent is
actually sadistic. (Nurse, 1964; Young, 1953). From her
study of 200 case histories and personally devised
questionnaires, Young concluded:

It is not the impetuoustiow of a harassed

parent nor even the transitory brutality of an

indifferent parent expressing with violence the

frustrations of life. It is not too severe

discipline . . .fiIt is the perverse fascination

with punishment as an entity in itself divorced

from discipline (Young, 1953, p.44).
She also found that the aggressive parent was abusive to-
ward all members of the family and tended to react
aggressively to both intrafamilial and extrafamilial crises.
Nurse (1964) confirmed Young's findings thatéabusive
parents were brutal to all members of the family including

the spouse. She reported that 13 of 20 abusive parents in
her study behaved aggressively toward their spouses. She

5
claimed that the aggressive parent tyrannically ruled the
home and used physical punishment as an outlet for hostile
feelings rather than for discipline.

Instead of aggressiveness being a salient person-
ality trait, Feinstein (1964) suggested that the abusive
parent's family situation in the present awakened conflicts
and resentments from his own childhood. During 80 hours of
group therapy, he studied six women with thoughts of in-
fanticide and concluded that the most prominent psycho-
logical factor underlying their infanticidal impulses was
their strong resentment towards their parents and mostly
toward their mothers. VThus, he conceived of child abuse as
a diaplacement of resentment from their own childhood to
their own children. This is somewhat similar to the con-
cept of role reversal.,

Bryant (1965) contended that there was more than /
one personality type associated with child abuse. From his
general observations, he inferred that there were really
three personality types: the aggressive type, the rigid
compulsive type who becomes frustrated by the child who
does not fulfill their expectation, and the passively
dependent type who competes with their children for attention.

The majority of research has focused on the inter-
action between social and psychological factors. (Beveral
studies have reported that severe marital difficulties
usually accompany child abuse (Elmer, 1965: Young, 1953;

6
Zalba, 1966). Elmer (1965) performed one of the few con-
trolled studies. lUsing 50 families in which child abuse
occurred, she discovered that~desertion, separation, and
divorce were more common in the abusive families.” Young
(1953) indicated.the fact that the home atmosphere was
hostile and conflictual. In a study using therapy inter-
views to obtain information from mothers who had committed
infanticide, the investigator learned that the act fre-
quently followed the actual or psychological abandonment
by husband or lover (Olive, 1965). After analyzing 80 case
histories of child abuse, Delsordo (1963) concluded that
~more than half of the cases of abuse arose from the pro-
jection of parental conflicts on to the children.
Some investigators concluded that child abuse was a

, reaction to a multitude of social stresses (Elmer, 1965;
Zalba, 1966). Elmer (1965) observed that besides having
severe marital difficulties, these parents were exposed to
the chnonic stresses of insufficient income, menial jobs,
and larger than average families. On the other hand,
Delsordo (1963) suggested that the overflow of life
frustration accounted for the abuse in less than 20% of his
80 cases. This evidence suggests that singly, this is not
a major determinant of child abuse.

~‘The rejection of the matQEEZi role has also been
considered an important area in the understanding of abuse.
Elmer (1965) compared 50 abusive mothers with a control

groUp. Using the PARI, he found that the abusive mothers

f. I If] [.1 I I... I: I III. III: II , I. 1‘ 1

7
demonstrated significantly greater rejection of the maternal
role. The abusive mothers were also more rejecting of
children in general. In his group therapy with women who
had infanticidal impulses, Feinstein (1964) discovered that
these women married passive effeminate men with hopes of
relinquishing the maternal obligations to them and of ex-
pecting these men to mother them. However, these implicit
arrangements were not accepted by the husbands, and the
women were forced into the maternal role. Similarly,
Gladston (1964) was impressed by the abusive mother's
masculinity and apparent distaste for rearing children.

- Reinhart (1967) contended that abusive mothers
actually did not differ from other mothers. He suggested
that child rearing was fraught with frustrations. Child
abuse occurred where there was a momentary lapse in the
mother's capacity for tolerating stress. He commented that
I'Perhaps we should be amazed that more children do not
suffer assault by parents who have come to their wit's end.'
(p.14).'

J’Blumberg (1964) attributed the mother's brutality
with her child to the lack of Opportunity for self-
expression or distraction which results from restrictions
of the maternal role.” They vent their frustration toward
the children and rationalize their aggressiveness by assign-
ing physical punishment as part of their parental prerogative.

Another point of view conceives of child abuse as

the product of a power struggle between parent and child.

8
Newson wrote “We find one important thing involved is self-
esteem between parent and child . . . The mother decides
she has got to show who's boss. Both she and the child
are afraid of losing face and in the end brute force settles
it (quoted in Blumberg, 1964, p.40).” Delsordo (1963)
found that this type of situation was rare in his sample
and was usually restricted to cases in which the child was
at least seven years of age.

Several other factors have been examined in relation
to child abuse. Zalba (1966), in his review of the
literature, stated that none of the studies have found
child abuse to be related to any particular socioeconomic,
ethnic, racial, or religious group. Another frequently en-
countered finding reported in the literature was the abusive
parent's social isolation (Elmer, 1965: Mdffifi'a Gould,
1963; Nurse, 1964; Young, 1953,1963: Zalba, 1966). Manber-
ship in outside organizations and recreational activities
was virtually nonexistent. {Because of their uneasiness with
people, they were known to have no close friendships and
little contact with relatives. The lack of emotional con-
tact with peOple made it more difficult to adjust to the
stresses of life. //

Thus, there are various hypotheses for child abuse’/
presented in the literature. Unfortunately, many of them
were derived from poorly designed, uncontrolled studies.
The purpose of the present investigation was to study the

psychological needs and frustrations of abusive mothers,

9
their perception of family life, their general attitudes
toward children, their sense of environmental restriction,
their feelings of self-reliance, their feelings of
belonging, and their capacity to nurture and administer to

their child's needs.

METHO D

§s consisted of 10 abusive mothers (AM's), those
women who severely abused their children, and 10 control
mothers (CM's). The latter were selected on the basis of
their similarity to the AM's in regard to socioeconomic
class, age, and education. The AM's were referred to Wayne
County Clinic for Child Study by physicians who from medical
evidence suspected abuse. Evidence of repeated abuse can
be detected by physicians through the use of xprays and by
the observation of bodily injuries (Caffey, 1946: Elmer,
1963; Fisher, 1958: Kempe, 35.91.,1962). Fractures at
different stages of healing, subdural hematomas, soft tissue
swellings, and skin bruising are some of the physical in-
dicators of abuse. The diagnosis is strengthened when the
type of injury is incongruent with the parent's description
of the accident and when observations of the child reveal
emotional unresponsiveness. Because of the careful screen-
ing of the AM's by the clinic's staff before they were
referred for testing, the probability is high that the vast
majority of them were guilty of child abuse. Nevertheless,
at the time of testing, the guilt or innocence of the parent
could not definitely be established. The sample of abusive

parents was limited by the criteria that the mother, not

10

11
the father, committed the abuse and that the child abused
was three years of age or less. These criteria may have
lead to the selection of a greater number of separated,
divorced, and unmarried women than would have been found in
the general population of AM's. Four of the AM's and two
CM's were either not married or separated from their
husbands at the time of testing.

The CM's were obtained through the cooperation of a
neighborhood social service organization at the Brewster
Housing Project in Detroit. The investigator was presented
with a list of mothers who lived in the Brewster Housing
Project. Those women who most nearly resembled the AM's
in regard to age, education, and socioeconomic class were
selected for the study.

The majority of both groups was comprised of lower
socioeconomic class Negroes. There were eight Negroes
among the AM's and ten among the CM's. The social class
groupings were based on an adaption of the system of
Hollingshead and Redlich (1958). Instead of using resi-
dence, total income was divided into seven categories and
served as one of the determinants of social class. (Refer
to Appendix A for exact income divisions). Data comparing
the AM's and CM's are summarized in Table 1.

Half of the AM's were tested at the Wayne County
Clinic for Child Study and the other half in their homes.
All CM's were tested in their own homes. Before testing,

the AM's were told only that the clinic would like to have

12

.c0eumsomum encumn Aconom moumcaEumu Hmsoa>aoca use» moamacmam Na u.wuoz

.oumspmum Hoonom seam emu.
.ooauuus uoz zz.

 

> H uu Ha mma un .22 «a cane
> a nu seem as uu .22 mm onus
> a nu as one nu ass as one
> a nu «a on nu .22 on mum
> H o oa me we as mm one
> m a a om on a an mum
> m o as om ea on em suo
>. H II m Oma II *2: an run
> n me as on mm m am one
> m an 0 mm an m am one
> m ma me nu me a ma mno
> m «a as oo as a mu mum
>H N 0H «a ma cu m om euo
>H H m as One mm m mm sum
> m m we om ON a ma muo
> N an as mm mm m mm mum
>H m scum NH ova as m mm «no
>H a as we ems as m a” «us
> m as an ace sanuuununnnvaa mm duo
> m as He was as (gas em Hum
musao ceuoafisu n.0cmnmsm m.umnuoz msoocH possum: unuam museum:
Hudson no .02 ceausosom waxes: c033 emm muse» ems gm
sauce

mmsoum aowucou use e>ansnm anon so sumo .a magma

13
some additional information from them in order to be able
to best help them. The CM's were contacted directly by the
investigator and asked if they would participate in a
research project on attitudes and feelings of young mothers.
All those women who were contacted agreed to participate in
the research. The total time required for testing varied
between one and a half to two and a half hours for each
person. -

Each g was presented with four different tests.
Firstly, they were presented with selected subscales from
The California Test of Personality (CTP) (Thorpe, g§.g;.,
1953). The five subscales used were Self-Reliance, Self-
Esteem, Sense of Personal Freedom, Feelings of Belonging,

and Withdrawing Tendencies. They were then administered

 

the Family Concept Inventory (FCI) (Palonen, 1966) and next
the Manifest Rejection Scale (MR) (Hurley, 1965). Lastly,
the Thematic Apperception Test (TAT) was given. The TAT
cards used were 1, 2, 3BM, 4, 6BM, 7BM, 7GF, BBM, 9GP, 12M
12F, 13B, 13MF, 17GF, lBGF. The TAT cards were presented
in the preceding order. .

(TThe CTP was selected for use in this study because
its subscales corresponded to certain personality attributes
thought, by some investigators, to be significantly related
to the dynamics of child abuse. Tindall (1955) found that
the CTP correlated beyond the .01 level of significance
with each of the scales of the Heston Personal Adjustment

Inventory, global ratings of adjustment of teachers and

l4

psychologists, sociometric ratings, The Rotter Incomplete
Sentence Test, and natural setting observations. Using the
CTP, Peak (1963) differentiated a group of children in the
ninth grade who were undergoing psychotherapy from a con-
trol group. Each of the subscales of the CTP consists of
15 Yes-No forced choice items. The main defect of the in-
strument appears to be its susceptibility to faking (Mathis,
1959).)

l, The FCI (Palonen, 1966) is a paper and pencil
format adaption of the Family Concept Q-Sort (Van Der Veen,
Huber, Jorgens, and Neja, 1964). The FCI requires the
endorsement of one of the following five reSponse categories:
strongly agree, tend to agree, neither agree nor disagree,
tend to disagree, strongly disagree. Van Der Veen, 33.31.
(1964) successfully differentiated predetermined high and
low adjustment families with the Family Concept Q-Bort.
This investigator also found that the Family Concept Q-Sort
correlated significantly with a 23 item adaption of Locke's
Test of Marital Adjushment. Using the FCI, Palonen (1966)
confirmed the latter finding.

The MR (Hurley, 1965) consists of 30 items which
deal with the parent's general disciplinary policies. The
reSponse categories are identical to those of the FCI.
Hurley (1965) found that the MR correlated substantially
with Eron's 24 item Punishment Index. The Punishment Index
focuses on the parent's responses to their children's

aggression. Also, Hurley found that the parent's MR scores

15
were significantly, but negatively, associated with their
children's IQ scores.

The TAT was introduced to the §,with the following
statements. “I am going to show you some pictures one at a
time. Using your imagination, make up a story by getting
an idea from the picture. Elaborate on your story as much
as possible. Each story you tell should have an ending,“
If the §_neglected to include an outcome to the story, the
examiner would inquire "How does the story end?“.
Pathogenic Scoring 2; Eng 25;

The TAT was analyzed by two different scoring
systems. The first scoring system was developed by Meyer
(1964, 1967) and elaborated upon by Mitchell (1965). This
system successfully differentiated between groups of mothers
of schizophrenic and non schizophrenic children. The pur-
pose of the scoring system was to help determine if these
mothers were able to satisfy their children's needs or if
their own needs took precedence over the needs of the child.
The general criteria for scoring was twofold. Firstly, if
there was an interaction between a dominant and dependent
individual with conflicting needs, the story was scorable,
and, if not, the story was unscorable. Secondly, if such
an interaction took place in a story and the dominant indi-
vidual took the dependent individual's needs into account,
the story was scored benign. If, on the other hand, the
dominant individual either did not meet the Specified needs

of the dependent individual or met his own needs at the

In I I ‘I In] I I! I'll" I'll (Ill ‘IIIIV II. II:

l6

eXpense of the other person, the story was pathogenic.
Mitchell (1965) empirically found that the predictive power
of the scoring system could be increased by considering
both pathogenic stories with benign outcomes and stories in
which the main character was permitted to escape a patho-
genic situation as benign. Each story was designated as
benign, pathogenic, or neutral. For each person, a total

.__A Pathogenic
ratio was computed: Pathogenic + Benign.

The assumption underlying this analysis of the'TAT
was that pathogenic stories reflect an inability of the
parent to meet the needs of his children or empathize with
them. Pathogenic stories also tended to reveal a lack of
basic trust in the environment's capacity to emotionally
satisfy one's needs. Thus, individual's with high patho-
genic ratings are probably peOple who feel that the
environment is cold and ungiving. Because of their felt
deprivation, they are unable to satisfy others.

Nggd and Neegygrustration Sporings for the TAT

The second scoring system attempted to evaluate
the individual's needs and frustrations. The needs scored
were dependency, independence, dominance, aggression,
nurturance, and affiliation.

Dependency was scored when the main character
revealed a need for sympathy, consoling, guidance, or help
from another character in the story, or when the main
character performed some action or sought a goal for the

stated purpose of earning praise, approval, or recognition.

17
Frustrated dependency was scored when the main character
expressed a need for help which was not given but was still
hOped for in the future, when the main character expressed
a need for affection or attention or bemoaned the fact that
he had been emotionally deprived, when the main character
mentioned being lonely, insecure, or depressed because of
the psychological or actual loss of some other figure in
the story, or when the main character did not carry on
after personal loss.

Independence was scored when the main character
eXpressed a desire for less imposed restrictions or for
escaping, when the main character acted independently in
his actions, such as complying with his own ideas or wishes
instead of following the wishes of another character, when
he acted contrary to orders, or when he carried on after
some personal loss. Frustrated independence consisted of
actions, such as the main character complying with the
demands of another character although he eXpressed the
desire to do the contrary, being placed in a situation
where he was unable to escape, having to act negatively to
his own wishes, being punished for independence, or being
placed in conflict concerning a wish to fulfill his own
desires or follow the wishes of another.

Dominance was scored when the main character
attempted to influence, persuade, or control other characters

in the story. It was also scored when other characters in

18
the story looked to the main character for leadership.
Frustrated dominance occurred whenever the main character's
attempts for dominance were unsuccessful.

Nurturance occurred when the main character con-
soled, sympathized, advised, protected, or helped others.
It was also scored when the main character expressed a
desire to help although he actually did not assist anyone
in the story, or when other characters in the story looked
to the main character for succor, support, or help. There
was no nurturance frustration score.

Positive affiliation was scored when a pleasant
interaction occurred between peOple of equal status, such
as friends or lovers, or when there was an expressed wish
for this type of positive interaction. Frustrated affiliation
concerned itself with negative interactions between peOple
in equal status relationships. A.boyfriend deserting his
girlfriend, quarreling between husband and wife, or stating
that the main character had no friends were examples of
frustrated affiliation.

Aggression, unlike the previous needs, was scored
regardless of whether or not the main character was the
instigator. It was scored whenever there was verbal or
physical aggression. This included criticism, scorn, blame,
hate, punishment, etc. Aggression also occurred both when
there was actual violence between the characters in a story
and when one of the characters was said to be angry without

acting out this emotioni)

19

(All needs, except aggression, are thus interpreted
from the point of view of the main character. If the main
character was passively acted upon by various environmental
presses, the need scored would be the complement of the
environmental press. For example, if the main character
was dominated by another character, frustrated independence
would be scored. If the environment nurtured or helped the
main character, this would be scored as dependency. Con-
versely, if the main character interacted with the environ-
ment, only the actions of the main character would be scored.
The reactions of the social environment, however, will help
determine whether the needs are frustrated or not.

The scoring system was quantified by the following
criteria: One point was given for the expression of a need.
Another point was added to the score if the need was elaborated
upon to a considerable degree or was involved in the main
theme of the story. A point was given if the need was ex-
pressed in the outcome of the story. No one need was given
more than three points per story.

Total frustration scores for each individual were
derived by the following alternative methods. If the need
was said to be frustrated in one part of the story but
satisfied in another part, the various independent expres-
sions of the need were separately scored as frustrated and
not frustrated. For example, if the main character was
forced to play the violin in the main part of the story but

independently decided not to play it in the conclusion, the

20
need in the main body of the story was scored as frustrated
independence but the independence need expressed in the
conclusion would not have been given a frustrated scoring.
On the other hand, if a need was eXpressed in the main
body of the story without any initial indication of whether
or not it would be frustrated until the outcome of the
story, then the need frustration scorings of the initial
expression of the need would depend on whether that need
was frustrated later in the story. All needs were not con-
sidered frustrated unless there was definite evidence that
this was the case.

A total accumulative score was derived for each
need and converted into percentages which were based upon
the total numerical equivalent of all the needs for that fiQ)
For example, if there were two dependency points and the
total for all the needs was 10 for a particular g, then his
total dependency score was I% or .20. Aggression was ex-
cluded from the calculation of the frustration score per-
centages because it had no frustration scoring. Greater
expression of aggression in the stories would tend to
decrease the frustration percentages and this would suggest
a contradiction to the notion that aggression and frustra-
tion were positively correlated to some degree.

All of the TAT stories were evaluated without
prior knowledge of whether the g was an AM or a CM. A
second judge was given the TAT stories of eight gs to re-

evaluate, four AM's and four CM's, and also without any

21
knowledge of the group status of the g, he scored the TAT
stories according to both scoring systems. The second
judge provided an indication of the relative objectivity
and reliability of the scoring systems.

RESULTS

The inter-scorer reliabilities for various analyses
of the TAT were obtained by means of productamoment
correlations. The §'s total score for each need was used
in the computations of the correlations. Each of the
reliabilities for the separate TAT need and need frustration
ratings was above .80. The average inter-scorer reliability
coefficient for the need ratings was .85 and for the need
frustration ratings .83. The reliability for the TAT
pathogenic index ratings was derived from each §fs total
pathogenic score. The inter-scorer reliability for the
pathogenic index was .94.

The grtest for small samples (Hays, 1963) was used
to compute the significance of all the between groups
statistics in the study. The small sample size, the
heterogenity of the sample, and the crudeness of the measur-
ing instruments were viewed as sufficient justification for
the use of the .10 level of significance in interpreting
the results.

Table 2 shows thatCthe AM's were differentiated
by their amount of self-esteem according to the CTP. The
AM's were significantly lower (p/;.02) on self-esteem.
Self-reliance, sense of personal freedom, feelings of belong-

ing, and withdrawing tendencies did not significantly

22

23

Table 2. Personality measure comparisons of the abusive

and control mothers

Means Standard t
Deviations values
Abu- Con- Abu- Con-
sive trol sive trol
California Test of Personality
Self-Reliance 9.2 11.0 3.52 2.36 1.24
Self-Esteem 8.7 12.0 3.77 1.91 2.56**
Sense of Personal Freedom 9.5 11.1 3.34 2.18 1.21
Sense of Belonging 10.5 12.4 4.00 1.22 1.27
Withdrawing Tendencies 8.8 10.6 4.64 2.45 1.05
Family Concept Inventory 91.4 130.5 49.20 26.90 2.09*
Manifest Rejection Scale 54.0 66.0 11.20 10.90 2.12**
TAT Ratings
Pathogenic 70.7 39.3 17.24 19.44 3.63***
Dependency Needs .27 .23 .18 .14 .59
Independence Needs .24 .26 .10 .09 .24
Affiliation Needs .11 .11 .06 .07 1.07
Aggression Needs .19 .13 .11 .13 .22
DOminance Needs .07 .08 .05 .02 .42
Nurturance Needs .05 .16 .04 .15 2.36**
Dependency Frustration .18 .07 .12 .09 2.17**
Independence Frustration .18 .15 .14 .08 .68
Affiliation Frustration .12 .09 .08 .08 .66
Dominance Frustration .06 .05 .05 .07 .31
Total Frustration .56 .38 .13 .15 2.73**

*pfi;.10 Using two tailed test of significance.

*fp<:.05 Using two tailed test of significance.
***p<;.002 Using two tailed test of significance.

24
differentiate the groups. The results reveal that the AM's
had significantly (9.4.10) poorer attitudes toward their
families. The obtained §_value (2.09) fell just short of

the .05 significance level (5 2.10). The CM's, on the

.05-
other hand, were significantly higher than the AM's on the
MR.(p¢:.05). Thus, the AM's registered less overtly reject-
ing attitudes toward their children than did the CM's. The
AM's showed a significantly higher (2<<.002) pathogenic
index than the CM's. There was high inter-scorer reliability
for the pathogenic ratings. The inter-scorer reliability,
using the product-moment correlation, was .94. The Table 2
results reveal that the groups could not be differentiated
on their dependency need scores, need for independence
scores, aggression, need for affiliation scores, or need
for dominance scores. The significant finding for the need scor—
ings was that the AM's showed a lesser (p (.05) need to
nurture. The AM's also differed reliably from the CM's on
dependency frustration (p<:.05) and total need frustration
(py:.02). Both of these measures had reasonably high inter-
scorer reliabilities. The product-moment correlation for
the TAT need ratings was .83 and for the need frustration
ratings was . 859

Pearson product-moment correlations were determined
among all the measures which significantly differentiated
the two groups (Table 3). (Significant inter-correlations
among the measures which differentiated the two groups were

the FCI and CTP Self-Esteem subscale (5;.78), TAT pathogenic

1|
I'll-ll
Ii
Pllli‘
I

".1

5 i I 0" ,- in,

25

Table 3. Product-moment correlations among statistically
significant measures

MR Pathogenic Self-Esteem Nurturance Dependency

Frustration
FCI -.12 -.58** .78** .25 -.39
MR —.24 .09 .26 -.40
Pathogenic -.43 -.62** .25
Self-Esteem .15 -.32
Nurturance .19

*fp .05 Using two-tailed test of significance.

index and FCI (gs-.58), and TAT pathogenic index and the
TAT nurturance scoring (EF'o52)- These findings provide
some confirmation of the underlying assumptions which were
stated concerning the psychological significance of the
pathogenic story€> Meyer (1964) and Mitchell (1965) assumed
that the pathogenic story revealed both the individual's
inability to administer to the needs of others and his per-
ception of the environment's unwillingness to meet his own
needs. The high negative correlation between pathogenic
ratings and nurturance supports the assumption that the
pathogenic story reflects an inability to nurture others.
The second assumption is confirmed by the significant
negative correlation between the FCI and the pathogenic
ratings. The FCI obtains information concerning the indivi-
dual's satisfaction with his family. Thus, @e FCI measures

the individual's perception of how well the family is

26

meeting his personal needs. The negative FCI-pathogenic
ratings correlation supports the assumption that the patho-
genic story reveals the perception of an ungiving environmenE)

(The high correlations between both the FCI and CTP
Self-Esteem subscale and the pathogenic ratings and CTP
Self—Esteem subscale (nearly significant at the .05 level)
suggests that self-esteem is associated with the perception
of the environment's unwillingness to gratify one's needs:)
and with the capacity to give of one's self to others.

A McQuitty Typal Analysis (1961) was performed onffix
the correlations in Table 3. The Typal Analysis is an
elementary type of factor analysis. Figure 1 shows the A
three sets of factors analyzed by the McQuitty Typal Analysisx
method. The three clusters were: 1) FCI-CTP Self-Esteem
subscale, 2) TAT pathogenic indexPTAT nurturance rating,

3) TAT dependency frustration ratings-MR. The Typal Analysis
revealed a fair degree of independence between the various
measuring instruments. Also, the FCI was found to inter-
correlate significantly with a greater number of the other
test measures than any other single measuring instrument,

and it accounted for the largest proportion of the total
variance. The pathogenic index had the second largest
number of significant intercorrelations and also accounted
for the next largest proportion of the total variance.)
Therefore, the FCI and pathogenic index appear to be the
most salient measures followed by the CTP Self-Esteem subscale

and the TAT dependency frustration ratings, reapectively. 3

[Initial

(MI. Ill ’
t
[I
I‘lv.
Illllll.
JIITn

III.
III‘.
In

Ill
IIIII
‘III

I!

27

TypalnAnalysis of personality measures which discriminated
abusive from control mothers

 

—{ -.5 )
/ F81

stagnant? 64562138 Beams?
rs.43)

    
 

 

r=. 39 )~— 1.11....

Fig. 1. Factors are represented by pairs of letters
with double arrows. The capital letters signify the measur-
ing instruments: AsFCI, B:MR, C=pathogenic index, D=CTP
Self-Esteem subscale, E=TAT nurturance rating, F=TAT depend-
ency frustration rating.

DI SCUSS ION

Interpretation of Findingg

The high pathogenic scores of the AM's suggest that
they have difficulty empathizing with or administering to
their children's needs. Also, the high pathogenic index of
the AM's indicates that they tend to perceive their own
emotional needs as being frustrated by the environment.
The AM's eXpressed greater frustration of their dependency
needs in their TAT stories. This fact suggests that these
women have strong unmet dependency needs which probably
originated from feelings of neglect or rejection during
their childhood. This finding is congruent with DeFrancis's
hypothesis that “Parents who abuse their children are parents
who react violently to their own unmet dependency needs.“
(DeFrancis, 1963, p.3). It can be Speculated that the in-
tensity of the unmet dependency needs of the AM's causes
them to be extremely narcissistic and consequently incapable
of identifying with their children. Administering to their
children's needs is probably felt to be a tremendous burden
because it conflicts with the parent's narcissism.

The AM's also have low self-esteem. The low esteem
appears to be related to their perception of the environ-
ment's unwillingness to meet their needs and the inability

of administering to the needs of others. This finding was

28

29
indicated by the high correlation between the pathogenic
index and CTP Self-Esteem. Thus lack of self acceptance
seems related to difficulty in accepting others. In
general, child acceptance has been found to correlate highly
with self acceptance (Mendinnus, 1963). These prior
reports were substantiated by the difference in self-esteem
between the AM's and the CM's in the present study.

The FCI data showed that the AM's have a negative
view of their families. This finding suggests that the
current family relationships are perceived as not suffi-
ciently meeting their emotional needs. Therefore, the
adequacy of the present family situation seems to be an
important factor in understanding the immediate stresses
involved in child abuse. This was indicated by the fact
that the FCI accounted for the largest pr0portion of the
total variance.

The hypothesis that child abuse is a reaction to a
multitude of social stresses (Elmer, 19667 Zalba, 1966) was
not supported by the present study. Some investigators
have felt that insufficient income, menial jobs, larger
than average families, and early marriages create the major
sources of social stress which result in the diSplacement
of frustration to the child.(:In the present investigation,
however, reliable personality differences were found between
the AM and CM groups deepite the fact that the study con-

trolled for social class, income, size of family, and age

30

when first married. Since the AM's showed distinctive
personality features, it would appear that child abuse is
not solely produced by social stress£>

(he higher MR scores of the CM's over the AM's was
uneXpected. This finding might be accounted for by a
defensive stance of the AM's.:>They had been requested by
The Wayne County Juvenile Court to come to The Wayne County
Clinic and undoubtedly were concerned about possibly
divulging information which might conceivably be used in
criminal proceedings against them. (gnother possible ex-
planation of the MR finding is that the AM's may have con-
‘siderable difficulty eXpressing their hostility toward their
children and may have consequently develOped the defense
'mechanisrcsof reaction formation and denial to help control
their aggressive impulses.:)This interpretation receives
some support from Megargee's study of murderers (1967). He
distinguished two types of assaultive murderers on the basis
of their inhibition level. The first is the undercontrolled
aggressive type who has low inhibitions for aggressive be-
havior. The second type, the chronically overcontrolled
type, is characterized by massive defenses against the ex-
pression, or even the awareness, of hostility. Thus, the
dynamics of the parents who abuse their children may,
according to the findings of the present study, correspond
to those of the latter type. This possibility is also con-

sistent with Komisaruk's view (1966) that the abusive

31

parents usually conceived themselves as ideal parents. He
interpreted this perception as a reaction formation against
their hostile feelings.
Unconfirmed Altegpate gypptheses Concerning Child Abuse

@he hypothesis that the abusive parent is generally
an overtly hostile, aggressive person was not supported by
this investigation. The AM's, as a group, did not express
more hostility in their TAT stories. However, it should be
pointed out that two of the ten AM's revealed a greater
amount of aggression in their stories than the other Ss:>
Probably, these two mothers were aggressive individuals.
PEEEEEEZL_EE:3,EEELESE~E'homogenous group and, as suggested

w
by.Megargee (1967), subtypes may be profitably identified.
W W

 

 

met-u"
. ' i...-

 

v
Generally, however, the AM' s are not typified by aggressive
assaultive women.

Also unsubstantiated by the present data was the

M____.,.._...-—-—-""" "’ “" 'u-u a... a,
“ M- w..._- .L ands-H... u- -- team-H- -- M.
u.‘

 

 

 

 

hypot thesis that child abuse arises fromuthe frustrations of
being restricted by maternal reaponsibilities. No reli-
-utmn. __~‘_-
ablgflbetwpgn:grpupsmdiffe5ences were found on either the
CTP Sense of Personal Freedom subscale or on TAT ratings of

the need for independence and independence frustration.

 

 

The TAT need for independence index was defined as a desire
for fewer fmposed restrictions, for escaping from a situ-
ation, or an expressed desire for acting independently.

Frustrated independence included situations in which the

.ua~b 7mm ‘1- 45‘me - '1.” film H'- “w,

main character felt trapped by circumstances, dominated by

Mfiw...”

external authority, in conflict about whether to act in-

dependently, or tied down to an undesirable situation.

.puu‘ 1‘!

In! nun-r ..

32
Since these indices did not differentiate between the two
groups of mothers, it appears that frustrated independence
is not a major determinant of child abuse.
Child abuse has been viewed as the product of a
power struggle between parent and child. Newson (in

"' *‘ow ‘ky—‘n

Blumberg, 1964) suggested that the mother feels compelled

 

.H. Manet—vi.» nu em-- 4.. “Aww, fl
”*"1 "Rh“-..‘tg “
Kn
‘M

to show the child Iwho isI boss and to use brute force to "M

,n- M‘

settle conflicts. In the present study, in order to in-

directly determine if the AM is overly authoritarianI and be-

flr-«flauumsww‘sW» VM-I M . “MN“,

comes overly frustrated by the child's willfulness, the

5‘ Lfl-‘Ilw A4

investigator scored the TAT for need for dominance and
dominance frustration. The present results indicate, how-

ever, that the need to dominateflwasmnot a differentiating

m‘w «MI—thi- rift

 

 

   

ea...‘

the results of this study may not be widely generalizable.

It would appear desirable to confirm the present findings

by employing a larger sample size and a wider distribution
of social class. Since the AM's would be facing criminal
proceedings and might be defensive about answering questions,
it would be advisable to include a measure of defgngiveness

Also, the present study demonstrated tIhatI family life is

an important variable in the dynamics of child abuse.

.....

‘ “M -

Future studies might also objectively assess the function-

ing of the family by means of interview ratings or

33
observational studies rather than being limited to only the
mother's perception of the family. This perception may
have been distorted to coincide with her own pessimistic
perception of the environment.

Further research is needed to determine the AM's

defenses and dynamics in dealing with aggressive impulses.

Megargee (1967) suggests that while one type of assaultive \.

m- '5‘“
py—wulfl“ .*1‘ J Lu»,

W

offender possesses massive defenses against the awareness.

‘W ‘
-..._+~ um..- -I-
rv"~r-‘

,vamu n-- “M
W

or expression of hostility, he may become momentarily

overwhelmed by his aggressive feelings. One possible

 

 

method of investigating the ability to eXpress hostility

f 40- -cmr U... Waugh“...

would be to instruct the S3 to relate aggressive stories

fr‘w ‘W‘I‘

to the TAT cards

REFERENCES

v”B1umberg: M. When parents hit out. Twentieth Centugx. V
1964, 173, 39-44.

'v/anrdman, H. A project to rescue Children from inflicted
injuries. Social Work, 1962, 1(1), 43-51.

Bryant, H. Physical abuse of children: an agency study.
Child Welfare, 1963, 33, 125-130.

Caffey, J. Multiple fractures in long bones of infants
suffering from chronic subdural hematoma.
American Journal 92 Roentgenology, 1946, §§(2),
163-173.

Curtiss, G. Violence breeds violence--perhaps. American
Journal 2; Psychiatry, 1963, 120, 386-387.

X'DeFrancis, V. Child abuse: preview of a nationwide survey.
Children's Division, The American Humane
Association, 1963.

t/’ Delsordo, J. Protective casework for abused children.
Children, 1963, ;Q(6), 213-218.

Elmer, E. Identification of abused children. Children,
1963, igfis), 180-184.

Elmer, E. The fifty families study. Published report of
the Public Health Service, National Institute of
Mental Health, 1965.

u ”Feinstein, H. Group therapy for mothers with infanticial
impulses. American Journal 2: Psychiatry, 1964,
120, 882-886.

, Plato, C. Parents who beat children. Saturday Evening
"' m, October 6, 1962, 235, 30-35.

V/Galdston, R. Observations on children who have been

b/ physically abused and their parents. American
Journal 9: Psychiatry, 1965, 122, 440-443.

34

35

Lx” Harlow, H.F., Harlow, M.K., Dodsworth, R.O., & Azling, G.L.
Maternal behavior of rhesus monkeys deprived of
mothering and peer associations in infancy.
Amggican Philosophical Society, 1966, 119(1), 58-66.

Hays, W.L. Statistics for psychology. New York: Holt,
Rinehart, & Winston, 1963.

Hollingshead, A.B., & Redlich, F.C. Social class and
x: mental illness: 3 community study. New Ybrk:
Wiley, 1958.

Hurley, J.R. Parental acceptance-rejection and children's
intelligence. .Merrill-Palmer Quarterly, 1965, 11,
19-31 0

Kaplan, M. Deaths of young studied by the city. N York
Times, May 5, 1962, 29.

 

L Kempe, H., Silvenman, F., Steele, B., Droegemueller, W., a
Silver, H. The battered-child syndrome. Journal
9; the American Medical Association, 1962, Lgl,
17-24.

L/// fi'Komisaruk, R. Clinical evaluation of families in which
child abuse has taken place. Symposium presented
at the Workshop on Child Abuse of the Wayne County
Juvenile Court, Detroit, 1966.

‘zfi Komisaruk, R. Clinical observations of the abused child
syndrome. Paper presented at the Conference on
Child Abuse at Wayne State University, Detroit,
June, 1965.

Lazarus, E.- The abused child: a new entity. Paper pre-
sented at the 12th. Annual Social Worker's Forum,
Detroit, April, 1966.

Mathis, C. Note on the susceptibility of the California
Test of Personality to faking. Psychological
Repgrts, 1959, g, 527.

’Medinnus, G.R., & Curtis, F.J. The relation between

maternal self-acceptance and child acceptance.
Journal 9; Consulting Psychology, 1963, 21(6),
542-542.

\.

legargee, E. Matricide patricide and the dynamics of
aggression. Symposium on murder within the family
presented at the American Psychological Association
Convention, Washington, D.C., September, 1962.

36

by means of the TAT. Unpublished masters
dissertation, Michigan State University, 1964.

/ Meyer, R. A study of the schiZOphrenogenic mother concept
(/

' L//Y:5M9Yero R., & Karon, B. The schiZOphrenogenic mother con-

cept and the TAT. ngchology, 1967, gg, 173-181.

genic mother concept by means of the TAT. Un-
published doctoral dissertation, Michigan State

' Mitchell, K. An elaboration of a study of the schiZOphreno-
V/‘ f

/ 1“

University, 1965..

Morris, 6., & Gould, R.W. Role reversal: a concept in
dealing with the neglected-battered child syndrome.
In the neglected-battered child s ndrome. New York
Child-Welfare League oTAmerTca,‘19"€3T 'p'p. 29-49. '

Nathan, H. Abused children. American Journal 9;
Psychiatgy, 1965, 122, 443.

Nurse, S. Familial patterns of parents who abuse their
children. Smith College Studies ip Social Work,
1964, pg, 11-25.

Palonen, D.J. Interpersonal perceptions and marital adjust-
ment. Unpublished masters thesis, Michigan State
University, 1966.

Peak, 8. California Test of Personality: a study of
validation. Doctoral thesis, Florida State Univer-
sity, 1963, Dissertation Abstracts. 23, 1281.

Reinhart, J. Child-beating laid to lost ideal. Newsday,
March 11, 1967.

Shaffer, H. Child abuse: search for remedies. Editorial.
Research Reports, 1965, 1(18), 343-359.

Scholoesser, P. The abused child. Bulletin pf the
Menninggr Clinic, 1964, 28, 260-268.

A'Steele, B., Pollock,, & Davoren, E. Observations on the

battered child syndrome and the law. For use at
the Washington, D.C. Conference on Child Abuse and
the Juvenile Court, May, 1966.

Thorpe, L.P., Clark, W.W., & Tiegs, E.W. Mgpggl for the
California Test 9; Personality. California Test
Bureau, 1953.

37

Tindall, R. Relationship among indices of adjustment
status. Educational and Psycholggical_Measurement,

1955, ,1__5_, 152-162.

Van Der Veen, F., Huebner, B., Jorgens, 8., & Neja, P.
Relationships between parents' concept of the
family and family adjustment. American Journal 9;

Orthopsychiatry, 1964, gg, 45-55.

J? Young, L. The behavior syndromes of parents who neglect
and abuse their children. Unpublished doctoral
dissertation, School of Social Work of Columbia

University, 1963.

 

44' Young, L. Wednesday's children: up study 9; child ngglect
and abuse. New York: McGraw-Hill, 1954.

 

APPENDIX A

The Seven Income Categories Used to Derive Social Class

Social Class Weekly Income Range
in Dollars

I Over 500

II 301-500

III 201-300

IV 161-200

V 101-160

VI 60-100
VII Under 60

38

39

mA

no
NwH

AA
0A

0A

mm

Ab

mm

MA
HA

HN

mh
mN
MA
MA
0A

m

mm fig
mm gm
H mVA mmA
MA NH
HA ¢A
MA AH
mH vA
0A mH
h o

mumnuoz Houucoo

om Vm

mv vm
gmH Am
AA 0A
MA MA
NA 0

NH 5
HA 0

m w

0A

Mb

OMA

AA

NA

AH
OH

m

#0

mm

5AA

NA
NA

NA
vA

00

mm
«A
MA

vA

MA
#H

mm #0

fin mv
H AmA vv

mH m

MA mH

OH 5
DA V

0H m

muomHssm

00H

Vh
00A
NA
MA
MA

mA
NH

m

mm

No
mm

0H

5

mumnuoz m>HmsQ4

mm

mm
hNH
MA
mH
mA

MA
mA

0

Nfl

ov
«NA
NA
NA
OH

NH
¢A

m

Hm

hv
«w

v

hm mm

mm mm
hNH Ne
CA N
AH N
HA m

m N

mm xoch
uHsmmonumm

00 m2

ow Hum

N anucmocmB
mcezmusnuwz
AH msHmsoAmm
no umcHHmom
h Eoommum
Assamumm
mo omsmm
m EwoummluHom
m mosmHHum
:uHmm
mac

musmsfluumcH
mcHHSmmmz

measumm oncomonuum ass was .ms .Hom .meu no um Hossfi>fi6aH non mmuoum 36m

m NHQmem4

mm.N Nv. NN. ho.H em. mm. 05Hm> u
mH. eo. No. no. no. mo. mH. HA. mo. 0A. «A. mH. coHu
IoH>oo

Gumccnum

MA. mo. mo. mo. HA. NH. MA. MA. 0N. VN. MN. 5N. G002

 

 

40

ms. mo. delIllmmdi mod :quI, 60:, mo. mm44 ~m4| o Amm¢ ANA
so. Ha. so. as. so. 0 ea. mm. em. ma. mm. mm. m
em. 0 mo. 0 ma. ea. ma. em. ma. mm. ms. 0H. m
so. 0 ea. mo. ea. ma. 0 NH. am. as. me. ma. a
mo. as. so. mo. ma. so. as. HA. mm. om. ms. mm. 6
HA. 0 0 mo. as. 60. mo. ma. mm. ma. mm. mm. m
mo. so. 0 0 mm. ma. mm. mm. mm. ma. ma. om. v
mm. mo. 0 mo. so. «A. 0 mo. mw. mo. me. we. m
ma. mo. ma. mo. 0 mo. mm. 6H. ma. mm. mm. mm. n
so. 0 ma. oa. eo. «a. Hm. mm. mm. mm. he. ea. H

.coo .mdnd .coo .msna .sOU .msn4 .coo .msnd .GOU .msnd .coo .nsnd mm

mosnusuusz mucosHsfia coHumHHHumd sonmoummd ouswosmmoocH mucoocmmmn

ass on» some sm>fluoo

mommucmuumm somz HouHmOHonu>mm mo cOmAHMQEoo

U NHQZHmm<

41

 

Mh.N HM. MM. mm. hH.N 05Am> u
MA. MA. no. Mo. mo. mo. mo. eA. mo. NA. coAu
InA>oo
numossum
MM. MM. M0. M0. mo. NA. MA. MA. 50. MA. smmz
gm. me. 0 AH. v0. mo. NA. 50. II. o 50. OH
MM. Mm. 0 NA. v0. 0 MA. MA. MA. MM. m
MN. Nv. Mo. 0H. mo. MN. no. mo. Mo. HA. m
Mv. Mb. vA. no. vH. vA. vH. NM. 0 wA. 5
mg. MM. mo. 0 5A. mo. NN. MN. 0 5H. 0
CM. MM. 0 MD. MA. MD. MA. OH. MA. vN. M
MM. MM. 0 0 ON. ON. MN. MA. MC. MA. v
MA. Ho. 0 0H. 0 mo. 0 0H. 00. vN. M
MM. No. MA. #0. 0 0A. HA. ¢H. MN. mm. N
me. am. mA. 6A. so. 6A. , mm. mm. o o A
.cou .502. .s00 .22. .200 .55. .cou 45¢ .sou 59¢ mm
coAumuumsum coAumuumsum coAumuumsum coAuouumsuh coAumuumoum
A6909 oucMCAEOQ soApnAHAuud mocmocmamocH Nucmcsmmmn

849 0:» scum po>AHmn

momoucooumm coHumuumsum Ummz mo semAummEou

Q Xanmmm¢

7"?8 1 Q m?)

HICHIGQN STRTE UNIV. LIBRQRIES

 

31293101280653