The Real Objective of Mendel's Paper

FLOYD V. MONAGHAN AND ALAIN F. CORCOS

Department of Botany and PlantPathology

Michigan State University
East Lansing, Michigan 48824, U.S.A.

ABSTRACT: According to the traditional account Mendel's paper on pea hybrids reported
a study of inheritance and its laws. Hence, Mendel came to be known as "The Father of
Genetics". This paper demonstrates that, in fact, Mendel's objective in his research was
finding the empirical laws which describe the formation of hybrids and the development of
their offspring over several generations. Having found these laws (and not the laws of
inheritance that he is generally credited with) he proposed a theoretical scheme involving
the formation of germinal and pollen cells in hybrids and their combination in fertilization
competent to explain why his laws took the form that they did. Mendel's research shows a
pattern of development closely paralleling the stages of empirical investigation beginning at
the level of qualitative description in common language rising through four levels of
increasing abstraction and culminating in a fifth level, his simple mechanical theory. This
mechanism met all the tests that a theory of this type must pass. In that respect Mendel was
highly successful and this success might be the reason why he did not push his theory to a
higher level, a sixth level involving the use of particulate determiners; despite this fact,
Mendel was credited with having taken exactly that step.

KEYWORDS: Mendel, theory of hybrids, Mendelian laws, symbolic notation, levels of

conceptualization.

Imagination ... which is ever wandering beyond the

bounds of truth, joined to self-love and that self-con-
fidence we are so apt to indulge prompts us to draw
conclusions which are not immediately derived from the
facts; . .. Hence it is by no means surprising that, in the
science of physics in general, men have so often formed
suppositions, instead of drawing conclusions. These
suppositions, handed down from one age to another
acquire additional weight from the authorities by which
they are supported, till at last they are received even by
men of genius, as fundamental truths.

A. L. Lavoisier
The Elements of Chemistry (1789)

INTRODUCTION

Traditionally, Mendel has been regarded as the father of the science of

Biology and Philosophy 5: 267-292, 1990.

© 1990 Kluwer Academic Publishers.Printedin the Netherlands.
268 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

genetics, the discoverer of the laws of heredity and the creator of a viable
particulate theory of inheritance. This traditional account of Mendel's
work, to use the name suggested by Olby (1979, p. 54), has been
challenged with little success since its inception (Corcos and Monaghan,
1990). The main challenge has been the view that Mendel's research was
not concerned with heredity but with the behavior of hybrids. Olby (1979)
championed that view in an outstanding paper entitled 'Mendel No
Mendelian'. More recently Callender (1988) also has proposed that
Mendel was not primarily concerned with developing a theory of inherit-
ance but with demonstrating "the existence of 'constant' hybrid forms
and [elucidating] their supposed role in the development of new species."
In this context Callender introduced an original but not convincingly
supported idea, that "Mendelism came into being, historically, as a sophis-
ticated form of the doctrine of Special Creation." Given this, Callender
concluded that one should see Mendel's papers on Pisum and Hieracium
as supporting "his [Mendel's] attachement" to this concept.
In this paper we shall show that the "traditional account" of Mendel's
work is wrong on three major points. First of all, examination of his
famous paper, Experiments on Plant Hybrids (Mendel, 1866), will show
that these experiments were not concerned with heredity and its laws but
were concerned with the formation and development of hybrids and the
laws related to this. Second, we shall show that, contrary to the traditional
account, Mendel was using empirical methods in designing and carrying
out his experiments. And third, careful examination of Mendel's Experi-
ments will show that he did not explain his results by employing invisible
particulate determiners, paired or otherwise. As a corollary of these points
we will show that the traditional Mendelian laws of segregation and
independent assortment are not given in the paper. Mendel was able,
without their use, to explain his results. The fact that he did not explain his
results by assuming the existence of particulate determiners is extremely
important, for this means that he did not develop the ultimate explanatory
scheme for which he has been given credit. Nevertheless his experiments
were carried out with such care and explained so clearly that others, at a
later time, looking from another point of view, were able to build on them
and come up with a workable theory of heredity.

A STUDY IN THE BEHAVIOR OF HYBRIDS

Part 1. The Evidence from Mendel's Paper

The first clue to the purpose of the experiments is in the title of the paper,
Versuche iiber Pflanzen-Hybriden (Experiments on Plant Hybrids). If the
focus of Mendel's interest in this research on peas were on the inheritance
 THE REAL OBJECTIVE OF MENDEL'S PAPER 269

of characters it seems likely that there would have been some reference to
this in the title. However, as is evident, neither the German word for
heredity nor for inheritance appear there. Nor, for that matter, do they
appear anywhere in the paper, as Olby pointed out some ten years ago
(1979, p. 65). In fact the content of the paper is consistent with the title as
an examination of it will show. For later clues we turn to the paper itself
and begin with a quotation from Mendel's introductory remarks.

Artificial fertilization undertaken on ornamental plants to obtain new color variants

initiated the experiments to be discussed here. The striking regularity with which the
same hybrid forms always reappeared whenever fertilization between like species took
place suggested further experiments whose task it was to follow the development of
hybrids in theirprogeny (Mendel, 1966, p. 1). (Emphasis added)

Here is a clear statement of what the observations were which initiated

the study and a first statement of his task as he saw it. Note that it was
the regular recurrence of the same hybrid forms which was striking, not
the regular reappearance of individual characters, not the regular way in
which certain parental characters are passed from parent to offspring.
Undoubtedly other hybridizers before Mendel had seen similar regular-
ities, but most of them did not regard them as of great consequence
(Roberts, 1965, Chapters 4, 5, 6). However, one of them, Gartner, did. He
wrote the following as translated by Olby (1979, p. 63)

... since very few hybrids show an equal mixing of the characters of both types, but
one factor in the union often preponderates over the other, so the question arises:
Which laws govern these modifications in the construction of hybrids? For these types
are not vague or the result of chance, on the contrary, they always arise in the same
manner and are of the same sorts.

In Mendel's case we know that he had spent two years at the University
of Vienna and that more than two thirds of his courses were in the
physical sciences. In these sciences such invariant relationships were
regarded as indicators that an underlying natural law was operating.
Mendel looked for such laws as they applied to hybrids, because he was
struck by "the regularity with which the same hybrid forms reappeared
... " This suggested to him that he should carry out "further experiments."
The following quotation indicates that he believed that there was a "...
law of the formation and development of hybrids ... " If Mendel had
not believed that such a law about hybrids existed, he would not have
searched for it.

That no generally applicable law of the formation and development of hybrids has yet
been successfully formulated can hardly astonish anyone who is acquainted with the ex-
tent of the task and who can appreciate the difficulties with which experiments of this
kind have to contend (Mendel, 1966, p. 21). (Emphasis added)
270 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

It is clear from this that the "generally applicable law" was not a law of
inheritance as such, but was a law "of the formation and development of
hybrids."
On the next page of the Introductory Remarks he criticized the work of
the ealier hybridizers and pointed out the route he intended to take in
trying to find the "generally applicable law".
Whoever surveys the work in this field will come to the conviction that among the
numerous experiments not one has been carried out to an extent or in a manner that
would make it possible to determine the number of different forms in which hybrid
progeny appear, permit classification of these forms in each generation with certainty
and ascertain their numerical interrelationships.This paper discusses the attempt at
such a detailed experiment (Mendel, 1966, p. 2). (Emphasis added)

Note that Mendel's criticism of the work of earlier experimenters was not
concerned with their failure to provide accurate descriptions of the
characters they were using or an accurate account of what happened to
them over several generations. These are the data he might have been
concerned about finding if he were preparing to study the inheritance of
the characters of traits. Instead he criticized their failure to provide data
on (1) what different forms appeared in the progeny of their hybrids, (2)
whether these forms appeared in each generation, (3) in what numbers
they appeared and (4) whether there were any patterns of number rela-
tionship among the various classes. These are the things he recognized that
he would have to obtain to discover that "generally applicable law"
relating to hybrids, not characters, if such a law existed. On p. 4, he further
clarified his task:
To discover the relationshipsof hybrid forms to each other and to theirparentaltypes it
seems necessary to observe without exception all members of the series of offspring in
each generation (Mendel, 1966, p. 4 ).
From the start, special attention was given to the Leguminosae because of their
particular floral structure. Experiments with several members of this family led to the
conclusion that the genus Pisum had the qualifications demanded to a sufficient degree.
Some quite distinctforms of this genus possess constant traits that are easily and reliably
distinguishable, and yield perfectly fertile hybrid offspring from reciprocal crosses.
(Emphasis added)

Here again we find Mendel concerned with the properties and behaviors
of the hybrids, not the behaviors of the individual traits and characters
composing them. He was only concerned that the traits be constant, easily
and reliably distinguishable and that the hybrid offspring of the crossings
be perfectly fertile.
Finally, in the section of the paper on Arrangement and Sequence of
Experiments, he stated the purpose of the experiments he proposed to do:
When two plants, constantly differing in one or several traits, are crossed, the traits they
have in common are transmitted unchanged to the hybrids and their progeny, as
 THE REAL OBJECTIVE OF MENDEL'S PAPER 271

numerous experiments have proven; a pair of differing traits, on the other hand, are
united in the hybrid to form a new trait, which usually is subject to changes in the
hybrid's progeny. It was the purpose of the experiment to observe these changes for each
pair of differing traits, and to deduce the law according to which they appear in suc-
cessive generations. Thus the study breaks up into just as many separate experiments as
there are constantly differing traits in the experimental plants (Mendel, 1966, p. 5).
(Emphasis added)

The new trait to which he referred is the "hybrid trait." And the law which
he will try to deduce is concerned with the changes in the hybrids in
successive generations.
Mendel then listed and discussed the various forms of peas selected for
his crossing experiments. Not all those forms could be used for the
reasons he pointed out:
However, some of the traits listed do not permit a definite and sharp separation, since
the difference rests on a "more or less" which is often difficult to define. Such traits
were not usable for individual experiments; these had to be limited to characteristics
which stand out clearly and decisively in the plants. The result should ultimately show
whether in hybrid unions the traits all observe concordant behavior, and whether one
can also make a decision about those traits which have minor significance in a classifi-
cation (Mendel, 1966, pp. 5 and 6).

As is evident, the traits were selected with the expectation that they
would all behave in the same way in "hybrid unions." He was able to
establish that his expectations were met. In each case the hybrid showed
exclusively the character of one of the two parents. Mendel called this
character the dominating character, whether that was a character of the
seed or pollen parent. The other parental character he called the recessive.
He also found that when the hybrids reproduced by self-fertilization, they
all behaved the same way, not breeding true but always splitting into the
dominating and recessive parental forms in a consistent ratio. This con-
stancy of behavior of the individual characters and of the hybrids was
essential to finding the law of hybrid development that he believed existed
and for which he was searching.
In the course of the monohybrid experiments Mendel established
empirically a series of invariant relationships, which can be extracted from
the text (not in Mendel's words, however). Although most of these
invariant relationships are in non-numerical terms, they have the character
of empirical laws. These are given below in serial order.
Law 1. The hybrid offspring of parents each true-breeding for one of
the contrasting characters of a trait are all alike and like one of the
parents. No intermediate types are formed.
Law 2. Reciprocal fertilizations yield the same hybrid forms. That is,
the hybrid trait will be that of the dominating parent regardless of whether
that be the seed parent or the pollen parent.
Law 3. When the hybrids are allowed to self-fertilize, the offspring (F2)
272 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

always appear in two classes, one like the hybrids of the first generation
(F,) and like one of the original true-breeding parents (the dominating)
and one class like the parental character not visible in the F, (the reces-
sive). No intermediate forms are produced. The two classes occur in an
approximate ratio of 3 dominating to 1 recessive.
Law 4. (a) when the F 2 recessives are allowed to self-fertilize they
always breed true. (b) When the dominating F2 are allowed to self-fertilize,
approximately 1/3 of them breed true while 2/3 of them behave exactly
like the F1 hybrids.
Mendel concluded his discussion of the second generation from the
hybrids (the F3) in the following way:

Since the members of the first generation originate directly from the seeds of the
hybrids, it now becomes apparent that of the seeds formed by the hybrids with one pair
of differing traits, one half again develop the hybrid form while the other half yield
plants that remain constant and receive the dominating and the recessive character in
equal shares (Mendel, 1966, p. 15). (Emphasis added)

Mendel expressed these relationships in symbolic form as follows:

If A denotes one of the two constant traits, for example, the dominating one, a the
recessive and Aa the hybrid form in which both are united, then the expression, A +
2Aa + a, gives the series for the progeny of plant hybrids in a pair of differing traits
(Mendel, 1966, p. 16).

There are two things about this formula which must be emphasized.
First the symbols do not represent genes, unit factors, or material particles
which determine characters (Heimans, 1971, p. 97). They do represent
observable charactersof plants and the expression 1 to 2 to 1 represents
the relative proportions of each class of progeny in the total series.
Second, the formula summarizes a large body of experimental results in
the form of an empirical law. It describes the proportions in which each
type of progeny is produced, but it does not explain why these progeny
are produced in these proportions. And Mendel made no attempt to
provide any explanatory mechanism at this point. It was enough to know
that the progeny were reliably produced in these classes and in these
proportions under well defined conditions.
Mendel thought that the pattern of development he had found in the
first and second generation of his hybrids could hold over many genera-
tions. To test this he carried some of the experiments through as many as
six generations from the hybrid. His results confirmed his expectations for
he found that in each generation the hybrid split up into hybrid and
constant forms in the ratio of 2:1:1. He summarized his results in a table
designed to show what happens to the hybrid in succeeding generations
under certain assumptions. Here, we clearly see that Mendel's center of
attention was on how the descendants of hybrids develop and split up, and
 THE REAL OBJECTIVE OF MENDEL'S PAPER 273

not on how parental characters were transmitted from one generation to

the next as would be the case if this were a study of inheritance. He began
with the observations of Gartner and K1lreuter and others on reversion in
hybrid offspring (Mendel, 1966, p. 16).

The observation made by Gartner, Kl1reuter, and others, that hybrids have a tendency
to revert to the parental forms, is also confirmed by the experiments discussed. It can
be shown that the numbers of hybrids derived from one fertilization decrease signifi-
cantly from generation to generation as compared to the number of newly constant
forms and their progeny, yet they can never disappear entirely. If one assumes, on the
average, equalfertility for all plants in all generations, and if one considers, furthermore,
that half of the seeds that each hybrid produces yield hybrids again while in the other
half the two traits become constant in equal proportions, then the numerical relation-
ships for the progeny in each generation follow from the Tabulation 1 below, where A
and a again denote the two parental traits and Aa the hybrid form. For brevity's sake
one may assume that in each generation each plant supplied only four seeds. (Emphasis
added)

TABLE 1

Expressed in terms of ratios

Generation A Aa a A:Aa:a

1 1 2 1 1:2:1
2 6 4 6 3:2:3
3 28 8 28 7:2:7
4 120 16 120 15:2:15
5 496 32 496 31:2:31
n (2" - 1):2:(2n - 1)

Note that the comparisons involve the numbers of the hybrids and that
of the parental plants in various circumstances, not the behavior of the
traits involved. The "newly" constant parental forms are mentioned only in
connection with their relative numbers in comparison to the hybrids. What
Mendel demonstrated was the power of his generalization about the way
hybrids split up in succeeding generations. Given this plus assumptions
about "equal fertility for all plants" and about equal productivity of each
plant (four seeds) he demonstrated theoretically that reversion to the two
parental types occurred in equal portions in each generation and that
although the hybrids represented an ever smaller proportion of the
progeny in each generation they did not disappear, but persisted in the
population.
The next section of the paper is concerned with "The offspring of
hybrids in which several different traits are associated." Again, the quota-
tion below indicates strongly that Mendel's focus in the paper is on the
hybrids and their behavior, not upon inheritance. Note particularly the
second paragraph (Mendel, 1966, p. 17).
274 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

In the experiments discussed above, plants were used which differed in only one
essential trait. The next task consisted in investigating whether the law of development
thus found would also apply to a pair of differing traits when several different charac-
teristics areunited in the hybrid throughfertilization.
The experiments demonstrated throughout that in such a case the hybrids always
resemble more closely that one of the two parental plants which possesses the greater
number of dominating traits ... Should one of the two parental types possess only
dominating traits, then the hybrid is hardly or not at all distinguishablefrom it. (Em-
phasis added)
When Mendel had completed his dihybrid experiments he summarized his
data by remarking that the forms found in the offspring of the dihybrids
could be classified into three groups: one in which all breed true in
following generations, one in which the members are true-breeding in one
trait and hybrid in the others (these vary in the next generation only for
the hybrid trait), and one in which the members are hybrid for both traits.
These last behaved, as he said, "exactly like the hybrid from which [they]
are descended." The members of these groups were in the ratio of 1:2:4
(Mendel, 1966, p. 19).
In the following symbolic notation the first four members of the series
constitute Group 1, the second four Group 2 and the remaining one
Group 3. Again, as before, the symbols denote characters not genes,
therefore, the combinations shown are not genotypes but specify the
characterstructures of each class ofprogeny

When, therefore, two kinds of differing traits[Seed parent A and B, Pollen parent a and
b] are combined in hybrids, the progeny develop according to the expression: AB + Ab
+ aB + ab + 2ABb + 2aBb + 2AaB + 2Aab + 4AaBb. (Emphasis added)
Indisputably this series is a combination series in which the two series for the traits A
and a, B and b are combined term by term. All the terms of the series are obtained
through a combination of the expressions: (Mendel, 1966, p. 20) (Emphasis added)
A + 2Aa +a
B + 2Bb +b

We should note that two members of the true breeding group (Group 1),
Ab and aB, display combinations of characters different from those
present in either parent and in the 3-trait cross where there are 8 true-
breeding types 6 of these 8 are new combinations. Thus in both series the
hybridization process has produced new stable, fertile, true breeding
forms.
These forms are probably those "intermediate forms" which are "non-
variant" to which Mendel called attention as "deserving special attention"
in his first letter to Nigeli (Mendel, 1966, p. 57).
In each of these classifications of offspring (for di- and trihybrids) the
basis of classification was whether the hybrid offspring were constant or
variable for certain of the traits. Since each of the parents was hybrid, this
echoes again his concern with whether hybrids breed true or are variable,
 THE REAL OBJECTIVE OF MENDEL'S PAPER 275

and what the laws of their behavior are, not with how characters are
inherited.
Two other examples will serve to support further the view that Mendel's
focus is on hybridization and not inheritance. The first is drawn from the
section of Mendel's paper on The Reproductive Cells of Hybrids. At this
point, he had completed a series of experiments designed to test certain
assumptions about the numbers and types of reproductive cells produced
by hybrids in particular cases. The data he had obtained matched very
closely the numbers and types of offspring he had predicted. He then
commented as follows:

Since the different constant forms are produced in a single plant, even in just a single
flower, it seems logical to conclude that in the ovaries of hybrids as many kinds of
germinal cells (germinal vesicles), and in the anthers as many kinds of pollen cells are
formed as there are possibilitiesfor constant combination forms and that these germinal
and pollen cells correspond in their internal make-up to the individual forms.
Indeed, it can be shown theoretically that this assumption would be entirely adequate
to explain the development of hybrids in separate generations if one could assume at the
same time that the different kinds of germinal and pollen cells of a hybrid are produced
on the average in equal numbers.
In order to test this hypothesis experimentally, the following experiments were
chosen (Mendel, 1966, p. 24). (Emphasis added)

Note that his hypothesis, which is not concerned with inheritance but
with the progeny of self-fertilized hybrids, consists of two parts: one
concerning the kinds of germinal and pollen cells produced by hybrids
and one concerned with their relative numbers. Note also that he is
concerned with the hypothesized process only in hybrids and not in true-
breeding types, and that no physical determiners (factors, pangens, or
genes) are involved in any way. These statements of Mendel about egg and
pollen cells are clearly hypothetical and do not correspond at all to the
statements that he calls laws.
It seems clear to a modern reader with a knowledge of genetics that the
formation of these different kinds of egg and pollen cells as proposed by
Mendel must, in the simplest case (monohybrids), involve segregation of
the characters. And, of course, in the case of polyhybrids, it now seems
clear to us that independent assortment of characters must also have taken
place. But, as we shall show in Part 2 of our paper, Mendel never
employed either of these concepts.
At the end of the experiments designed to test his hypotheses he wrote:

Thus experimentation also justifies the assumption that pea hybrids form germinal and
pollen cells that in their composition correspond in equal numbers to all the constant
forms resulting from the combination of traits united through fertilization.
The difference of forms among the progeny of hybrids, as well as the ratios in which
they are observed, find an adequate explanation in the principle just deduced (Mendel,
1966, p. 29).
276 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

Note that it is the difference of forms among the progeny of hybrids as

well as the ratios in which they are observed that "find an adequate
explanation in the principle just deduced" and not the pattern of inherit-
ance of particular traits.
The importance of this point for Mendel is shown by his emphasizing it
again in the final summary paragraph of this section in his paper. He had
just completed his demonstration of the series of progeny for hybrids in
which three traits are united.

The law of combination of differing traits according to which hybrid development

proceeds thus finds its basis and explanation in the proven statement that hybrids
produce germinal and pollen cells that correspond in equal numbers to all the constant
forms resulting from the combination of traits united through fertilization (Mendel,
1966, p. 32).

These examples drawn from Mendel's paper are sufficient, we believe,

to indicate clearly that Mendel's language throughout the paper is the
language of a student of hybridization and hybrids, their properties,
stability or variability, sterility or fertility, their behaviors, breeding true or
splitting up and reorganizing. Inevitably he worked with characters and
traits as the components of his hybrids. From time to time he made some
brief statements about their behaviors. However, it is not these statements
which are developed and extended. He touches upon these matters mainly
in passing and as a sidelight to his central concerns with the properties
and behaviors of the various hybrids and the laws which describe those
behaviors.
In conclusion the view that Mendel's work as being about hybrids, the
laws of their formation and of the development of their offspring, is well
supported by the evidence of Mendel's own words in his 1866 paper,
while the traditional view that it is about inheritance is not. Mendel's
statements of the purpose of his experiments and of the procedures he
intended to follow show that his attention was focused upon finding the
quantitative empirical laws which describe the formation of hybrids and
their progeny. It is our view that finding these laws was an essential step in
the attainment of Mendel's ultimate objective in this research, the creation
of a mathematically precise science of hybrids modeled on the pattern of
the physical sciences.
There are two points which must be made as clear as possible and
cannot be overemphasized. The first has to do with the nature of the
experiments themselves. In this matter it is important to recognize that
whether Mendel's experiments were devised to provide data for finding
the laws of hybrid development as we believe they were, or for finding the
laws of inheritance, they would necessarily involve the formation of
hybrids and their offspring. And this inevitably involves the transfer of
characters from parents to offspring. No matter which of the two purposes
was the one leading to Mendel's experiments, the experiments themselves
 THE REAL OBJECTIVE OF MENDEL'S PAPER 277

would have been essentially the same. The crucial difference would lie in
the reason for doing them and the consequent analysis and the interpreta-
tion of the data produced.
The second point is concerned with what we do not claim, that Mendel
was exclusively interested in hybridization and had no interest whatever in
inheritance. Although Mendel's paper and his research appear to us to be
focused entirely on hybridization and hybrid behavior and not upon
inheritance, it seems very unlikely that in the environment in which he
lived and worked Mendel would have been totally disinterested in the
problem of inheritance. This especially in view of Abbot Napp's expressed
concern in this matter (Wood and Orel, 1983, p. 67) as well as other
members in the community and of the agricultural and scientific societies
of which he was an active member. There is one small bit of evidence
from his second letter to Nigeli (Mendel, 1866-1874) in which he makes
a direct reference to inheritance, but only as something which did not take
place.

Part 2. The Scientific Method of Mendel

It is generally accepted that an investigation of the sort Mendel conceived

and carried out must be a response to a guiding hypothesis without which
there would be no basis for selecting experiments or organizing them in a
sequence or judging the worth of the information accumulated. As Lenzen
puts it (1969, p. 238): "Significant experiments require the guidance of
hypotheses which serve to predict the results of observation." Mendel
placed his hypothesis in the first two paragraphs of his Introductory
Remarks.
Mendel's working hypothesis appears to have been the result of some
experiments performed earlier than those on peas. In the course of these
experiments he was struck by the "... regularity with which the same
hybrid forms always reappeared whenever fertilization between like species
took place .. .". It appears that these observations led him to formulate his
working hypothesis that there must be ". . . a generally applicable law of
the formation and development of hybrids ... " The argument is of the
form that regularity of behavior implies the existence of an underlying law
to which the observed regularity is a response. Mendel would have been
familiar with this concept from his study of the physical sciences. In those
sciences, following the direction suggested by such clues had led to finding
many laws of great correlative and predictive value. The search for order
expressed in laws became the raison d'etre of his experiments with peas.
By the end of his experiments he had found such order expressed in the
law for Pisum. This is the way he described it.
The progeny of hybrids in which several essentially different traits are united represent
the terms of a combination series in which the series for each pair of differing traits are
combined. (Mendel, 1966, p. 22)
278 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

From his knowledge of how quantitative empirical laws had been

discovered in the physical sciences together with his experience with
artificial fertilization in flowering plants and his knowledge of the work of
the 18th and 19th Century plant hybridizers he decided that to obtain the
data needed for finding the law he was seeking, certain conditions must be
met. The experiments must be designed so that it would be ". . . possible
to [1] determine the number of different forms in which hybrid progeny
appear, [2] permit classification of these forms in each generation with
certainty, and [3] ascertain their numerical relationships."
What Mendel proposed to do, and did, was to extend the mathematical
and physical conceptions with which he was familiar from his background
in the physical sciences into an area where they had not been applied
before. In doing this Mendel was attempting to bring the study of hybrids
into line with contemporary practice in the more powerful and productive
physical sciences. As Mertz pointed out:

It [modern science] is based upon numbering and calculating - in short upon mathe-
matical processes; and the progress of science depends as much upon introducing
mathematical notions into subjects which are apparently non-mathematical, as upon the
extension of mathematical methods and conceptions themselves. (Mertz, 1896, p. 30)

When Mendel started his work he did not have a preconceived theory
of the formation of hybrids which he intended to test. He was functioning
as "an empirical worker" as himself said. The evidence for this can be
found not only in his 1866 paper but also in his correspondence with the
Swiss botanist Carl Nageli. In a letter dated April 18, 1867, the year after
his Experiments were published, Mendel wrote:
Permit me to state that as an empirical worker I must define constancy of type as the
retention of a character during the period of observation. (Emphasis added)

And a bit further in the same letter he wrote,

Therefore, 2Aa + A + a or A + 2Aa + a is the empiricalsimple series for two differing

traits (characters). Likewise it was shown in an empirical manner that if two or three
differing traits are combined in the hybrid, the series is a combination of two or three
simple series. (Emphasis added)

These statements of his own view of his method of working are clearly
consistent with the view derived from examination of his report.
When we examine the organization of Mendel's Experiments as he
described them in his 1866 paper, we find the sections reflecting a clear
pattern of development closely paralleling the stages of empirical investi-
gation. Mendel's account begins in common language of low technical
content and rises gradually through four levels of increasing abstraction as
indicated below and culminates in level five, the level of theory formation.
1. The level of qualitative description in natural language.
 THE REAL OBJECTIVE OF MENDEL'S PAPER 279

2. The level of quantitative description.

3. The level of invariant relationship partly expressed verbally and
partly quantitatively.
4. The level of empirical laws expressed verbally and symbolically.
5. The level of theory formation in which he proposed and tested a
theory of hybridization based upon a simple mechanism of gamete
formation and their union in fertilization.
We turn now to an examination of the Experiments using the levels of
discourse of conceptualization or of abstraction as a framework.

Level 1

The first four sections of Mendel's paper (Introductory Remarks, Selection

of the Experimental Plants, Arrangement and Sequence of Experiments
and, The Form of Hybrids) consist mainly of qualitative descriptions in
natural language. The single exception to the verbal description occurs in
the third section. In this he gives a very brief quantitative statement of the
number of fertilizations and of plants involved in each of the first seven
experiments. In the fourth section he returns again to qualitative, purely
verbal description of his material.

Levels 2 and 3

When we come to the next three sections of the paper (The First Genera-
tion of Hybrids, The Second Generation of Hybrids and Subsequent
Generations from Hybrids) we find Mendel working at Levels 2 and 3.
Here he is shifting from purely verbal description of his experiments and
results to a mixed mode, less qualitative and more quantitative and giving
emphasis to invariant relationships.
Mendel's way of studying hybrids differed greatly from that of most of
his precursors. He looked at the contrasting characters of individual traits
rather than the total ensemble of characters as his precursors did. In this
latter approach the hybrid was usually described in terms of its general
similarities to the father or mother plant or as intermediate between them.
The result was more or less vague descriptions of a "more or less" nature
whereas Mendel's results, on the contrary, were clear-cut. In order to
study the relationship of the various forms of progeny from a hybrid
Mendel resorted to counting the members of the individual classes of
offspring. Cohen and Nagel (1934, p. 282) comment that, "in many
enquiries counting the individuals who possess a certain character is the
only possible method of avoiding vague ideas." Thus the counting of the
numbers of the various classes of progeny offered Mendel a way to
sharpen distinctions and to determine numerical relationships among the
classes. Experiment 1 offers a good example.
280 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

Experiment 1. Seed shape. From 253 hybrids 7324 seeds were obtained in the second
experimental year. Of them 5474 were round or roundish and 1850 angular or
wrinkled. This gives the ratio 2.96:1.

Mendel continues in this fashion through all seven experiments and

then offers the following:
When the results of all experiments are summarized, the average ratio between the
number of forms with the dominating trait and those with the recessive one is 2.98:1
or 3:1.

This 3:1 ratio of dominating to recessive forms holds constant across all
pairs used in the experiments and is, therefore, within the limits of experi-
mental error, another invariant relationship, a law which can be expressed
as: The progeny of a cross, Aa X Aa, are produced in the ratio of 3 of the
dominating form to 1 of the recessive form.
Mendel then points out that "the dominating trait can have a double
significance here - namely either that of the parental characteristic or that
of the hybrid character." Which of the two it is for each individual case
can be determined only by producing the next generation, then classifying
and counting the progeny. Mendel did this for each of the 7 experiments
and found for each case essentially the same pattern as for the first
generation. He gave his results in Section 7, The Second Generation from
Hybrids.
Among 665 plants raised from the round seeds of the first generation 193 yielded only
round seeds and therefore remained constant in this trait; 372, however, produced both
round and angular seeds in the proportion of 3:1. Therefore, the number of hybrids
compared to that of the constant forms is as 1.93:1.

This ratio, of course, rounds to 2 to 1, i.e., 2 dominating forms that were

hybrids to each dominating form that was true breeding. When these
results were combined with the observed behavior of the recessives it was
clear that, on the average, for each trait, one half of the progeny were
hybrids and one half of the individuals were true breeding, divided equally
between dominating and recessive forms. And he comments on the
behavior of the recessives as follows:
Those forms that receive the recessive character in the first generation do not vary
further in the second with respect to this trait; they remain constant in their progeny.

Mendel has thus determined that the 3:1 ratio in which the dominating
and recessive characters appear in the first generation from the hybrids
"resolves itself into the ratio of 1:2:1 in all experiments ... " This invariant
relationship or law can be expressed as follows: The progeny of a cross,
Aa X Aa, are produced in the proportion of 1/4 true-breeding dominating
forms to 1/2 hybrid dominating forms and 1/4 true-breeding recessive
forms.
 THE REAL OBJECTIVE OF MENDEL'S PAPER 281

Level 4

In Section 7, when Mendel had carried some of the seven experiments

through from four to six generations "without any deviation becoming
apparent" he then generalized the 1:2:1 relationship by writing it in an
abstract letter symbolism:

If A denotes one of two constant traits, for example, the dominating one, a the
recessive, and Aa the hybrid form in which both are united, then the expression
A + 2Aa + a

gives the series for the progeny of plants hybrid in a pair of differing traits.

This is the law of development for the progeny of the hybrids showing
contrasting characters of a single trait. It is the first of the combination
series laws.
At this point we have reached an example of Level 4, the level of
empirical laws expressed verbally and symbolically. Mendel followed this
and concluded the section with the example given earlier, on pp. 9 and 10,
showing how this generalized formula together with two assumptions can
explain quantitatively the reversion of hybrids to their parental strains
while a few of the hybrids always remain. Although reversion had been
known qualitatively for a long time this was the first time it had been dealt
with quantitatively and deductively.
In Section 8 of his paper Mendel reported his study of "The offspring
of hybrids in which several differing traits are associated. Working at
Level 4 he generalized the results for the progeny of hybrids for any two
or three of his seven traits by writing them in his abstract letter symbolism.
His generalized formula for two traits is given on p. 14 of this paper. He
then related this to the single-trait formula above by stating that the more
complex formula could be obtained by the term-by-term combination of
two of the single-trait formulae A + 2Aa + a and B + 2Bb + b. In a
similar way the three-trait formula could be obtained by combination of
three single-trait formulae. Mendel concludes Section 8 with a verbal
statement of these complex relations (Mendel, 1966, p. 22):

The progeny of hybrids in which several essentially different traits are united represent
the terms of a combination series in which the series for each pair of differing traits are
combined.

Level 5

In the first eight sections of his paper Mendel had accumulated the data
and found the empirical laws derived from them. In Section 9 he turned to
282 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

providing a mechanism that would explain why the laws take the form that
they did (Mendel, 1966, pp. 23-32).
In each case where the formation of the progeny of hybrids was the
result of self-fertilization, the process always involved the union of egg
cells and pollen cells produced in the same plant. Mendel, therefore,
devised a series of experiments intended, as he wrote, "to throw light on
the composition of seed and pollen cells in hybrids." He had established
that the formation of hybrids and their progeny conformed to certain laws.
Since these processes always involved the union of egg and pollen cells it
followed that the processes of the formation of these cells and their union
must also conform to some law or laws.
In each of these prior experiments the clue leading to the law he had
sought was a consistent quantitative pattern of behavior. He followed the
same route in this case. In this instance he wrote that "an important clue"
was the fact that in Pisum constant [true-breeding] forms appear among
the progeny of hybrids and that they do so in all combinations of the
associated traits (Mendel, 1966, p. 24). He drew a second clue from his
knowledge of the work of earlier investigators in this area. Of this he
wrote,

In our experience we find everywhere confirmation that constant progeny can be

formed only when germinal cells and fertilizing pollen are alike, both endowed with the
potential for creating identical individuals, as in normal fertilization of pure strains.
Therefore, we must consider it inevitable that in a hybrid plant also identical factors are
acting together in the production of constant forms. (Mendel, 1966, p. 44)

There are some difficulties in understanding this passage because it is

easy for us to interpret identical factors as identical factors inside the cells,
but Mendel seems to have meant identical germinal and pollen cells.
Perhaps a clearer notion of the central meaning of the passage can be
obtained if we omit the phrase, "both endowed, etc." and also substitute a
different translation for the German word for factors "facktoren" in the
following way, with the omitted phrase shown by three dots, and the
substitution in square brackets [ ].

In our experience we find everywhere confirmation that constant progeny can be

formed only when germinal and pollen cells are alike ... , as in the normal fertilization
of pure strains. Therefore we must consider it inevitable that in a hybrid plant also
identical [germinal and pollen cells] are acting together in the production of constant
forms.

These changes place the argument squarely in the form that the same
effect is produced in each case by the same process or cause, i.e., union of
identical germinal and pollen cells. Here Mendel has stayed close to his
empirical data and drawn a conclusion from them without going beyond
Level 5.
 THE REAL OBJECTIVE OF MENDEL'S PAPER 283

We have summarized in Table form some of the relevant data available

to Mendel from the experiments performed thus far (See line 1 through 5,
Table 2).

TABLE 2

1. Number of traits followed 1 2 3

2. Number of contrasting characters 2 4 6
3. Symbolic notation for characters A,a A,a, B,b A,a,B,b,C,c
4. Number of constant forms in the progeny of 2 4 8
hybrids
5. Character structure of each constant form A,a AB, Ab ABC, ABc,
(from the breeding data) aB, ab AbC, Abc
aBC, aBc,
abC, abc
6. Types of egg and pollen cells produced A,a AB, Ab ABC, ABc,
aB, ab AbC, Abc
aBC, aBc,
abC, abc

In each case, comparison of the character structure of the constant

forms and the contrasting characters in the hybrid showed that the
character structure of each constant form corresponded to a possible
combination of the contrasting characters present in the hybrid parent.
And indeed it showed that the total series of progeny corresponded to all
the possible combinations of the parental characters. Since each of these
constant combinations in the progeny must have resulted from the union
of an egg and a pollen cell, these data opened a way to discovering the
"composition of these cells".
But the processes of formation of egg and pollen cells in a plant are not
subject to direct observation. Therefore, Mendel did not know what was in
the cells which endowed them with the potential for creating offspring of
certain kinds and thus made them either a type A or type a egg or pollen
cell. And for his purpose this was not necessary. It was only necessary that
they be formed in appropriate kinds and, on the average, in equal
numbers. Here is Mendel's statement:

Since the different constant forms are produced in a single plant, even in just a single
flower, it seems logical to conclude that in the ovaries of hybrids as many kinds of
germinal cells (germinal vesicles), and in the anthers as many kinds of pollen cells are
formed as there are possibilitiesfor constant combination forms and that these germinal
and pollen cells correspond in their internalmake up to the individualforms.
Indeed, it can be shown theoretically that this assumption would be entirely ade-
quate to explain the development of hybrids in separate generations if one could
assume at the same time that the different kinds of germinal and pollen cells of a hybrid
are produced on the average in equal numbers (Mendel, 1966, p. 24).
284 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

Given these assumptions Mendel could offer an explanation of why the

progeny of monohybrids were produced in three types, true-breeding type
A plants, true-breeding type a plants and hybrid plants of type Aa.
When the pollen and germinal cells about which Mendel wrote united
in fertilization they combined in all possible combinations as he showed in
the following diagram (Mendel, 1966, p. 30):

(1) (2) (3) (4)

Pollen cells A A a a

Germinal cells A A a a

Combination 1 (a type A germinal cell fertilized by a type A pollen

cell) gave the true-breeding type A progeny while Combination 4 (a type a
germinal cell fertilized by a type a pollen cell) gave the true-breeding type
a offspring. Combinations 2 and 3 each gave the hybrid type Aa offspring.
And the relative proportions of the progeny were correct, 1A + 2Aa +
1a. Not only did the process explain the recovery of the parental charac-
ters entering into the hybrid but it also accounted for the continued
production of hybrid offspring even though the parental hybrids were not
themselves true-breeding.
In regard to this Mendel wrote:

Thus repeated hybridization takes place. The striking phenomenon, that hybrids are
able to produce, in addition to the two parental types, progeny that resemble them-
selves is thus explained (Mendel, 1966, p. 30).

As a result of this line of reasoning we are able to derive the data

shown in line 6 of Table 2. This shows the types of egg and pollen cells
produced by the hybrids of various kinds.
An important point must be noted in connection with these results.
This is that by this demonstration Mendel was able to determine what
types of egg and pollen cells must be produced to give the progeny of
monohybrids which he had already obtained experimentally. The deduc-
tion takes the following form: if the progeny contains true-breeding type A
individuals, these must have been produced by the union of egg and pollen
cells of the same type, type A. Therefore, the hybrid parent must produce
type A pollen and egg cells. In the same way he established that they must
produce type a pollen and egg cells.
This led him directly to his "proven hypothesis" that in the hybrid, first,
there are as many kinds of egg cells and as many kinds of pollen cells
produced as there are constant forms in the progeny, second, that these
cells correspond in their internal composition to that of the constant
forms, third, that the different kinds of egg and pollen cells of a hybrid are
 THE REAL OBJECTIVE OF MENDEL'S PAPER 285

produced on the average in equal numbers, and fourth, that in fertilization

the types of cells unite at random in all possible combinations. "In order to
test these hypotheses experimentally" a series of experiments was carried
out. In each experiment Mendel predicted not only the forms which
should appear in the progeny but their relative numbers. This series
included monohybrid, dihybrid and trihybrid parent plants. In each ex-
periment he reported that "the proposed hypothesis finds adequate con-
firmation." He concluded this section of his experiments with the following
statement.

The law of combination of differing traits according to which hybrid development

proceeds thus finds its basis and explanation in the proven statement that hybrids
produce germinal and pollen cells that correspond in equal numbers to all the constant
forms resulting from the combinations of traits united through fertilization (Medel,
1966, p. 32).

(Note: The law to which Mendel refers was given on p. 27 above.)

In this example we have seen how Mendel was able to account for the
formation of the progeny of the monohybrid cross Aa X Aa - A + 2Aa
+ a without involving the existence of specific particulate determiners in
any of the cells, i.e., without developing any higher level of explanation.
This is important, because at the time Mendel was conducting his Experi-
ments, the most developed, most sophisticated of the sciences was physics,
followed closely by chemistry. Both of these sciences had developed a
sixth level of explanation involving invisible particles assumed to have
certain properties and relationships and assumed to be moved by invisible
forces. The use of such concepts embodied in the atomic theory and the
kinetic molecular theory had begun to tie together part of the body of
known empirical laws in these areas. In contrast the biological sciences
had only just begun to move in this direction in the early years of the 19th
century. Mendel was aware of the more sophisticated methods of explana-
tion used in physics and chemistry as a result of two years of advanced
study at the University of Vienna. However, he stopped short of using it in
his own research.
Although Mendel never reached the sixth level of abstraction in his
investigation of pea hybrids, he does approach it in his concluding re-
marks. There he refers to "cell elements, to antagonistic elements and to
the material composition and arrangement of the elements that attained a
viable union in the cell," in fertilization. He hypothesized that the union of
antagonistic elements is temporary in the variable hybrids with which
he worked, or permanent in true-breeding hybrids that other workers
reported. He also proposed that in cell reproduction "all cell elements
participate freely and fully and only those that differ separate from each
other." While the term element may suggest by association the chemical
286 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

elements with their component atoms which could be combined and then
set free again, there are so few similarities that the term cell component
would be as informative without carrying the connotations that the term
element does.
Mendel's statement above about cell elements is as near as he ever
came to proposing that antagonistic elements segregate in gamete forma-
tion. If he had done so this would still have been only a very limited
conception, for in his paper there is no suggestion of pairing of such
elements in hybrids or in true-breeding progeny. He was unwilling to go
beyond the assertion that "The distinguishing traits of two plants can, after
all, be caused only by differences in the composition and grouping of the
elements existing in dynamic interaction in their primordial cells." It is
surprising, given his strong empirical orientation that he is willing to
speculate this much. Since he felt that the Law for Pisum, which was based
on solid empirical data needed further confirmation it is not surprising
that at the best he felt that his whole speculation about cell elements is was
not supported by "well substantiated data."
We have stated in the first part of our paper that one cannot find in
Mendel's paper the well known Mendelian laws: the law of segregation
and the law of independent assortment. The reason we do not find these
laws even in an embryonic form is not evident at first. However, we are
now in a position to account for this.

The Law of Segregation

The usual form of this law states that there are paired entities called genes
present in each individual cell and that these genes or alleles, separate
from one another in gamete formation with each gamete receiving one
gene of a pair. Obviously the law cannot exist in this form in Mendel's
paper because the necessary concepts and relations were not developed
until many years later. If the law could not exist in that form, could it exist
there in a precursor form employing concepts and relationships available
to Mendel? And if so, where is the earliest point at which it might occur?
Since Mendel framed his empirical laws in terms of observable charac-
ters he might have formulated a law of segregation, or separation, of
characters at some point. If this were to be derived from observation the
earliest point at which he had pairs of contrasting characters available was
in his monohybrids. And he demonstrated clearly that while such hybrids
showed only the dominant character, the recessive character contributed
by one of the parents had not been lost because it reappeared in the
progeny of the hybrids. While he reported that the progeny contained
dominant and recessive forms in the ratio of 3:1, he did not write that
the characters present in the hybrid segregated and recombined in the
production of the progeny.
 THE REAL OBJECTIVE OF MENDEL'S PAPER 287

Thus, the first Mendelian law, the law of segregation is not present
anywhere in Mendel's paper. That it cannot be found has been said many
times by quite a few writers (Corcos and Monaghan, 1990) but they offer
little or no direct evidence bearing on this. Among these writers is
Callender (1988, p. 52) who offers the view that the law of segregation
"has for the best part of the Century served to obliterate the distinction
that Mendel clearly and repeatedly drew between two main groups of
hybrid plant forms: 'variable hybrids' in which segregation takes place and
constant hybrids in which segregation does not take place. But, a critical
reading of Mendel's paper reveals that Mendel did not make such a
distinction and this for a good reason, he did not find any constant
hybrids. To support his view Callender quotes the following paragraph
written by Mendel:
If two plants which differ constantly in one or several characters be crossed, numerous
experiments have demonstrated that the common characters are transmitted unchanged
to the hybrids and their progeny; but each pair of differentiating characters, on the
other hand, unite in the hybrid to form a new character, which in the progeny of the
hybrid is usually variable. The object of the experiment was to observe these variations
in the case of each pair of differentiating characters and to deduce the law according to
which they appear in successive generations. (Callender, 1988, p. 51)

He adds: "At the outset of his report [first part of Section 31 therefore
Mendel draws a distinction between hybrids whose progeny is variable -
that is to say, give rise to different forms in subsequent generations - and
those which do not."
This statement of Callender is wrong. The distinction that Mendel
makes is between characters of the parents of the hybrid, those that are
common to both and those that are not. He is not talking about hybrids
whose progeny is variable and those whose progeny is similar to the
hybrid.
When Mendel uses the term hybrid he is referring to the cross of two
plants which differ in the contrasting characters of one or more traits but
which are the same in the characters of other traits. These latter traits and
only these are transferred unchanged from parent to offspring in succes-
sive generations. But, when the characters are unlike in the two parents,
only one of the characters, the dominating is expressed in the hybrid.
However, both reappear unchanged in the offspring of the hybrid. In
this case the hybrid is variable, not true-breeding with respect to that
particular trait. For example, if Mendel had made a cross between the
following two plants AB X Ab, the progeny of the cross would have the
character structure ABb, using Mendel's notation. According to this the
resulting plant of the cross would be true-breeding for the first trait and
hybrid for the second. This example could be extended to show the hybrid
constant in one trait and variable in several other traits, e.g., ABbCcDdEe
or the reverse constant in more than one trait and variable in only one,
288 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

e.g., ABCDEe. This is very different from Callender's interpretation. He

sees Mendel's hybrids as totally true-breeding or totally hybrid.
If the law of segregation is not in Mendel's paper, what then of the
second "Mendelian" law, the law of Independent Assortment?

The Law of IndependentAssortment

Let us begin by quoting the law of independent assortment in the form in

which we usually encounter it. It states that the genes of each pair assort to
the gametes independently. Thus, in the case of a dihybrid such as the one
Mendel employed, the usual modern notation would be AaGg where A
symbolizes the round form of the pea and a the angular. And in the
second pair G symbolizes the yellow color of the seed and g the green
form. When gametes are being formed the A gene will segregate from the
a. The same thing will happen with the G and g genes. This is in accord-
ance with the law of segregation of genes.
When the genes are distributed to the gametes the G gene will assort to
one gamete and the g gene to another gamete. But when the Aa pair
assorts to these same gametes this process will be unaffected by the
assortment of the Gg pair and vice versa. Thus the A gene may appear in
a gamete paired with a G producing a gamete AG, or it may appear in a
gamete paired with a g producing a gamete Ag. In like fashion the a gene
may appear in a gamete paired with the G gene (aG) or with the g gene
(ag). Since under these conditions the pairing is random there will be an
equal numbers of all four couplings. This is the well known Mendelian law
of Independent Assortment. We have written this out completely so that
readers not familiar with the meaning of this law will understand clearly
the salient points of this section.
The key to Mendel's explanation of his data from di- and trihybrids
without using independent assortment was the hypothesis that there are as
many kinds of egg cells and as many kinds of pollen cells as there are
constant forms in the progeny of the cross producing them, and that the
"composition" of those egg and pollen cells corresponds to those of the
constant forms. Mendel was able to show that the character structures of
the remaining progeny of the cross also could be produced by combina-
tions of these few types of egg and pollen cells and that the proper number
of each was produced. He then had a "complete" explanation of the
formation of cells of the progeny in the series.
He was able to deduce the types of egg and pollen cells which were
needed to produce the observed character structure of a given offspring.
He could also deduce what combinations of characters would be "poten-
tially" in the egg and pollen cells from his generalization relating to
number of characters in the parents of the hybrid to constant form in the
progeny of the hybrid.
 THE REAL OBJECTIVE OF MENDEL'S PAPER 289

It is therefore very clear that Mendel arrived at the composition of the

egg and pollen cells in both the dihybrids and trihybrids without formu-
lating the concept of independent assortment. Strangely enough, the laws
he did not find, segregation, and independent assortment, which were the
foundation stones of genetics, were created from his data by others who
said he had discovered them. His major law for Pisum, the combination
series law, which he did discover, was clearly of little or no interest or
value to other hybridizers or students of inheritance either then or later.
And this law did not lead to the discovery of any other laws of the same
type in the work of other investigators. Indeed, when Mendel's paper was
rediscovered in 1900, his combination-series law was ignored by both de
Vries and Correns. Instead, the new emphasis on segregation imparted by
de Vries and Correns was to prove decisive for future developments and
future interpretations of Mendel's experiments. However, these did not
lead to law of independent assortment until 1913 when Thomas Morgan
first stated it (1913).

Part 3. Summary and ConcludingRemarks

In this paper we have presented a development of Olby's suggestion that

Mendel's paper and the research it described were concerned with the
behavior of hybrids and not with the inheritance of characters or traits per
se. In it we have endeavored to show that the real objective of Mendel's
work was the creation of a mathematically precise science of hybridization
modeled upon the physical sciences.
In this process we have not seen Mendel as being guided by any
preexisting theory about hybridization or about inheritance as suggested
by Fisher (1936). Instead we have come to regard him as being guided by
certain basic philosophical ideas derived from his interests and educational
background. These leading ideas can be summarized as follows:
1. Observed order and pattern in natural phenomena are evidence of the
operation of empirical laws.
2. Such laws exist in the biological world as well as in the physical world.
3. Empirical laws are best sought for in experiments yielding quantitative
relationships which can be formulated in symbolic terms and inter-
preted through existing mathematical principles.
4. Since the outcome of such experiments cannot always be predicted, it
is necessary to be guided in further experiments by such results as well
as by the over-all design of the experiments as originally conceived.
5. Given the empirical laws that state in concise quantitative, symbolic
form what happens, an attempt must be made to provide a mechanism
or process that will "explain why the laws are as they are and not other
wise." The explanation must draw upon and be consistent with con-
temporary knowledge in the field.
290 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

Mendel's work began with the designing and carrying out of a series of
experiments which provided the quantitative data that became the basis
for a series of empirical laws describing the formation of hybrids and their
properties over several generations. These laws were expressed in numeri-
cal-symbolic form. Using them he "explained" in quantitative form several
long known but hitherto unexplained observations related to hybrids.
Having obtained the laws he proposed an hypothetical mechanism to
explain why the laws were as they were. The mechanism consisted of two
parts, one concerned with the number and kinds of egg and pollen cells
formed by a plant and a second concerned with the manner of their
combination in fertilization.
On the basis of his hypotheses he proposed several experiments and
made quantitative predictions of the results which should be obtained. In
each case the results were "in complete agreement" with the predictions.
These results, in his view, provided a satisfactory confirmation of the
correctness of his conceptions. It is this system of laws and the culminating
hypotheses that provide the structure of what we have come to regard as
Mendel's Theory of Hybrids.
At this point Mendel's theory had met the test of internal logical
consistency and the test of reliable quantitative prediction within the limits
of the domain of behavior as Mendel has defined it. And it had enabled
him to explain quantitatively such phenomena as reversion and trans-
formation. In it Mendel had defined and demonstrated new ways of
experimentation by adapting physical methods of experimental design and
of analysis and interpretation of data to a biological problem. And he
had demonstrated that the law found for Pisum had at least limited
applicability to other kinds of plants. Within the limitations indicated
Mendel had every reason to feel that his quantitative theory of hybrids
formation and development was as complete as he could make it.
While he offered some speculations in his concluding remarks attempt-
ing to push his theory to a higher level of generality we have no evidence
that he worked intensively at this. Since he did not, he never developed
what might be called the sixth level of abstraction. This sixth level would
have been the proposal that the reason why a certain cell was of certain
type, A or a, was that particulate determiners were present in the cells and
that these determiners occurred in pairs in all cells of a plant except the
germinal and pollen cells in which they occurred singly.
A suggestion as to why he did not reach the sixth level may be found in
the following quotations from Willey (1955, p. 14).
Considered as a psychological event, an explanation may be described as a change in the
quality of our response toward an object, or an idea. This change is typically a release
from some sort of tension as in the ordinary cases of 'clearing up' any mystery ....
An explained thing, except for very resolute thinkers, is almost inevitably explained
away.
 THE REAL OBJECTIVE OF MENDEL'S PAPER 291

Given these conceptions we might say that when Mendel had proposed
and tested his theory and found that it worked, the tension which had
sustained his efforts for eight years was released. If he had not found his
theory complete and satisfactory he most probably would have continued
to think seriously about developing a more satisfactory one. As it was he
did not do so.
Although his interpretation seemed complete and satisfying to him,
Mendel was aware that it was not properly appreciated by his contem-
poraries. This may have been because they did not understand Mendel's
explanation, or if they did understand it they did not regard it as really
satisfactory. Perhaps it was too abstract or mathematical and not
sufficiently biological. At any rate, Mendel was sure that in the future,
sooner or later, he would be understood and that "his time would come".
Little could he suspect that when this did happen he would be appreciated
for something he had not done.
In 1865 Mendel saw his work as answering his question, what are the
laws of formation of hybrids and the development of their offspring over
several generations. In 1900 when Mendel's work was again brought to
the attention of biologists it was seen, not as answering Mendel's question,
but as providing an answer to a question that F. C. Napp asked in 1840
(Wood and Orel, 1983, p. 67), What is inherited and How it is inherited?
Thus, Mendel's data were reinterpreted as being about heredity and in this
context significant additions were made to it by researchers such as de
Vries, Correns, Bateson and others. Over a period of time these additions
resulted in the creation of what came to be known as Mendelian Genetics.
We have here then, a case of multiple interpretations of a given set of
data. In regard to this Franke asks (1957):
Are the general statements of science uniquely determined, or can the same set of
common sense experiences give rise to different general statements?

Here the answer is clearly that they are not always uniquely determined,
for it is evident that Mendel's data did "give rise to different general state-
ments," to those we are examining here and also to the more traditional
ones relating to heredity. It is interesting to note that although the mode of
explanation of the data has undergone multiple changes the data still
stand. The facts of observation have a different order of stability and
permanence than the explanations.
And so, with his breeding experiments, his emphasis on quantitative
data, his empirical laws and his proposed mechanisms of gamete formation
and fertilization Mendel layed all the essential groundwork for the creation
of Mendelian Genetics by later workers but was not, in the direct sense,
the Father of Genetics in the classical form.
We might say of these events that Mendel appeared on the stage in
1865 carrying his manuscript, spoke the prologue and left the stage. After
292 FLOYD V. MONAGHAN AND ALAIN F. CORCOS

a long interval the curtain rose in 1900 on the play, written in the
meantime by others, with another cast of characters and about a different
problem.

REFERENCES

Callender, L. A.: 1988, 'Gregor Mendel: An Opponent of Descent with Modification', Hist.
of Science 26, 41-75.
Cohen, M. and E. Nagel: 1934, An Introduction to Logic and Scientific Method, Harcourt,
Brace and Co., New York
Corcos, A. and F. Monaghan: 1990, Mendel's Work and Its Rediscovery: A New Perspec-
tive, critical reviews in Plant Sciences, Boca Raton, Florida (in press).
Fisher, R.: 1936, 'Was Mendel's Work Been Rediscovered?', An. of Science 1, 115-137.
Franke, P.: 1957, Philosophyof Science, Prentice Hall Series, Englewood Cliffs, N.J.
Heimans, J.: 1971, 'Mendel's Ideas on the Nature of Hereditary Characters', Folia Men-
deliana 6, 91-98.
Lenzen, V.: 1969, 'Procedures of Empirical Science', in O. Neurath, R. Carnap, and C.
Morris (eds.), Foundationsof the Unity of Science, Vol. 1, University of Chicago Press,
Chicago.
Mendel, G.: 1866, 'Versuche uber Pflanzen-Hybriden', Verhandlungen des naturforschenden
Vereines in Brunn 4, 3-47.
Mendel, G.: 1966, English translation of the Versuche, in Curt Stern and Eva Sherwood
(eds.), The Origin of Genetics. A Mendel Source Book, W. H. Freeman, San Francisco,
pp. 1- 5 5 .
Mendel, G.: 1866-1874, Letters to Ngeli as translated by Pitternick L. K. and A.
Pitternick, in Curt Stern and Eva Sherwood (eds.), The Origin of Genetics. A Mendel
Source Book, W. H. Freeman, San Francisco, pp. 56-102.
Merz, J. T.: 1986, European Thought in the Nineteenth Century, Blackwood and Sons,
Edinburgh and London, I, p. 30.
Morgan, T.: 191 3, Heredity and Sex, Columbia University Press, N.Y.
Olby, R.: 1979, 'Mendel no Mendelian', Hist. Sci. XVII, 57-72, as an appendix in Olby's
Origins of Mendelism, 2nd ed., University of Chicago Press, Chicago, 1985.
Orel, V. and A. Matalova: 1983, Gregor Mendel and the Foundation of Genetics, a
symposium, published by the Mendelanium of the Moravian Museum, Brno, Czechos-
lovakia.
Roberts, H.: 1965, PlantHybridization Before Mendel, Hafner, N.Y.
Willey, B.: 1955, The Seventeenth Century Background, Doubleday Anchor Book, Garden
City, New York.
Wood, R. J. and V. Orel: 1983, 'The Sheep Breeders' Legacy to Gregor Mendel', in Gregor
Mendel and the Foundation of Genetics, a symposium edited by V. Orel and A.
Matalova, published by the Mendelianum of the Moravian Museum, Brno, Czechoslo-
vakia.