GUYER 
DEFICIENCIES OF CHROMOSOME 
THEORY OF HEREDITY ^ublwljrh by it^ Mmfaraity nf ©inritmaii 
Series II SEPTEMBER-OCTOBER, 1909 Vol. V, No. 3 
Deficiencies of the Chromosome Theory 
of Heredity 
By MICHAEL FREDERIC GUYER, Ph. D. 
PROFESSOR OF ZOOLOGY, UNIVERSITY OF CINCINNATI 
ISSUED BI-MONTHEY FROM THE 
UNIVERSITY PRESS, CINCINNATI, O. published by the 
UNI VER SIT Y^ CINCINNATI 
Deficiencies of the Chromosome Theory 
of Heredity 
By MICHAEL FREDERIC GUYER, Ph. D. 
PROFESSOR OF ZOOLOGY, UNIVERSITY OF CINCINNATI 
Series II, Vol. V, No. 3 
September - October, 1909 
Issued Bi-Monthly 
from the 
BURNET WOODS • CINCINNATI- Deficiencies of the Chromosome Theory of Heredity 
I should like in the present paper to discuss the evidence 
upon which inordinate importance has been attributed to the 
chromosomes as vehicles of heredity, and to inquire into how 
far this theory agrees or fails to agree with some of the known 
facts of protoplasmic activity and of heredity. By chromo- 
some theory is meant, in this connection, the original theory which 
regards the various morphological parts of an adult as specifi- 
cally predetermined by corresponding anticipatory units which 
reside in the chromosomes of germ-cells. While many present- 
day workers, I am aware, do not hold rigidly to this concep- 
tion, nevertheless, for purposes of criticism it seems best to adhere 
to a consideration of the theory as expressed and still largely 
maintained by its founders. To get at the matter satisfactorily 
it will be best to examine, point by point, the .evidence upon 
which the theory is founded. 
/. Roux's ('83) dictum that mitosis is meaningless unless 
it is for the precise halving of qualities serially arranged, so 
that, qualitatively, each daughter-cell will resemble the other, 
is historically, perhaps, the first specification of the chromo- 
somes as centers of series of qualities. 
As the case then stood, no evidence had been adduced to 
show that chromatin is qualitatively differentiated, but theo- 
retical considerations, based chiefly upon the phenomena of 
what is commonly termed particulate inheritance, had led 
several biological workers to the assumption that a multitude 
of particles bearing incipient hereditary qualities must exist 
in the germ cells. In mitosis was found a mechanism seem- 
ingly designed for halving a series of differential particles; 
consequently, to the chromatin masses which make up the in- 
* Read before Section IV (Cytology and Heredity) of the Seventh International 
Zoological Congress, Boston, August 27, 1907. 
3 dividual chromosomes was assigned the role of being the actual 
bearers of the so-called hereditary qualities. As to Roux's 
avowal that, unless we postulate in the chromosomes discrete 
qualitative particles which require to be accurately halved, 
direct division would suffice and mitosis would be superfluous, 
we know that direct division does sometimes suffice, even for 
the production of cells which ultimately give rise to germ-cells. 
In 1894 Meves showed that apparently normal development 
of spermatozoa may follow amitosis in the salamander. More 
recently Child ('07) has brought forward what appears to be 
unmistakable proof of the fact in Cestodes. These cases, 
of course, neither prove nor disprove the qualitative nature of 
chromatin, but if chromatin is qualitatively differentiated, 
the results of these investigators show that the qualities can 
be distributed without the elaborate structures seen in mitosis, 
and Roux's argument, therefore, based as it is upon the sig- 
nificance of mitosis, is invalid. 
2. The fact that the nucleus is apparently an important 
center of chemical synthesis has been construed as a proof 
that it is therefore the vehicle of heredity. 
Judging from the results of physiological chemists, it seems 
probable that the synthetic chemical changes wrought by the 
nucleus result in part in the construction of incitive or acti- 
vating materials, in part, presumably, in the further elaboration 
of nutritive substances; but, to my knowledge, there is no evi- 
dence which will warrant us in assuming that the nucleus bears 
or makes, as it were, self-sufficient morphological units which 
at just the right time pass out and take up their proper position 
in the cytoplasm and with the more or less passive cooperation 
of the latter expand into the structures required. 
Everyone will admit, I think, that the cytoplasm of a 
given species of plant or animal is distinctive of that species 
in all cells of the organism, and there is no conceivable reason 
why it, any more than the nucleus, must be made so anew in 
the germ-cells of each generation. So distinctive of its kind is a 
given protoplasm, in fact, that there is great probability that in 
higher animals digested proteids, either in transit through the 
digestive epithelium or immediately upon entering the blood, are 
synthesized into a new proteid specifically characteristic of the 
4 animal in which it is present. Here, then, we have a degree of 
specificity initiated before the proteid in question has ever 
become cellular in nature. 
Of all cells in the body, those which show nuclei of enormous 
size in proportion to their cytoplasm are not cells which have 
a large number of morphological structures to establish, but 
gland-cells characterized by intense secretory activity in which 
chiefly one kind of product is elaborated. 
The death of denucleated, and the regeneration of nucleated, 
fragments of protozoa are phenomena sometimes cited as dem- 
onstrating the fact that the nucleus is the center of morphological 
synthesis. But may they not indicate only that some general 
substance elaborated or controlled by the nucleus is wanting? 
Loeb ('99) has shown that possibly in such cases failure to re- 
generate may be due largely to a lack of oxygen, and that the 
nucleus is apparently the chief source of the oxygen supply of 
the cell. Spitzer ('97) has established the fact, moreover, that 
certain "oxidation ferments" are nuclear in origin. In these 
experiments on protozoa the rapidity of regeneration seems to 
be in direct proportion to the amount of nuclear matter left in 
the piece-a question of quantity rather than quality. Indeed, 
if the nucleus is an aggregate of qualitatively different morpho- 
logical units, one would expect parts to be missing in the regen- 
erated protozoa in proportion to the amount of nuclear matter 
removed, but the evidence does not bear this out. The regen- 
eration is seemingly complete, only a longer time is required if 
but a small fragment of the nucleus is left in the piece. 
3. Another evidence that is commonly cited may be ex- 
pressed as follows: Offspring inherit equally from each parent; 
only one spermatozoon enters the egg normally, and this sper- 
matozoon, though exceedingly smaller than the egg, contributes 
the same number of chromosomes that are present in the egg. 
Since only the chromosomes are contributed equally by sperm 
and egg-cell to the fertilized egg, it is argued that they must be 
the physical bearers of the equal contributions of qualities from 
the respective parents. 
However, I think we may legitimately question the alleged 
fact of parental equality in inheritance. While it is true, ap- 
parently, that individual or specific characters may be derived 
5 equally from either parent, the fundamental characters that go 
to make up an organism-such, for example, as make it an animal 
and a vertebrate of a given family or genus-are characters that 
are common to both parents; and from a consideration of adult 
forms alone, therefore, we have no means of definitely determin- 
ing whether such characters are transmitted by one or by both 
parent germ-cells. When we come to consider embryological 
stages, however, the evidence at our command, meager though 
it be, plainly indicates that we do not inherit equally from each, 
but more from the maternal parent. 
Since I have discussed this matter at some length in a 
former paper (Guyer, '07), I shall cite but a single instance in 
this connection. Both Conklin ('05) and Lillie ('06), if I do 
not misinterpret their results, find that in certain forms, before 
the spermatozoon enters, the fundamental organology is al- 
ready roughly indicated in the egg by the distribution of so- 
called "formative substances.'' The egg then passes on to 
cleavage without waiting for a duplicate set of "formative sub- 
stances'' to pass out from the male pronucleus. Since there 
is evidently no biparental mechanism present in these early and 
more fundamental stages, we are not justified in maintaining 
that offspring inherit equally from each parent. Consequently, 
if we insist on restricting all of the mechanism of heredity to 
the chromosomes, the argument from the striking parity be- 
tween the chromosomes of male and female germ-cells loses its 
force. 
It is true that in these cases cleavage is not inaugurated 
until the spermatozoon enters the egg, but here we must dis- 
criminate between simply the stimulating effect of the sperm 
and its capacity for conveying hereditary qualities. That this 
is a valid distinction is evidenced by the facts of artificial par- 
thenogenesis, where by various artificial means the mere stimulus 
to cleavage may be brought about. 
It may be urged that although "organ-forming substances'' 
are present in the cytoplasm of certain eggs before fertilization, 
these substances have originated from the nucleus during the 
earlier history of the egg-cell. There is good evidence, in fact, 
to show that the cytoplasm of the egg is more homogeneous in 
appearance in earlier stages than later, and that substances are 
ejected into it from the nucleus. However, this does not prove 
6 that these nuclear emanations are morphological entities, and I 
see no sufficient reason for not regarding them simply as nuclear 
products which cooperate with cytoplasmic substance of equal 
importance in the construction of materials which may serve as 
a basis for further differentiation. In any event, the disparity 
between maternal and paternal inheritance would remain. 
Inasmuch as the bulk of the egg, even after fertilization, is 
cytoplasm chiefly of maternal origin, it is clear that the develop- 
ing organism is more maternal than paternal in its derivation. 
Nevertheless, we can see how the veneer of individual traits 
may be equally of maternal and paternal origin if, to express 
it crudely, we look upon cytoplasm and chromatin respectively 
as responsive mechanism and inciting agent-the character of 
the response depending upon both the constitution of the cyto- 
plasm and the materials (enzymes? nutritive substances?) 
emanating from the nucleus-subject, of course, to the restric- 
tive influences of the environment. We know, furthermore, that 
individual parental characteristics appear for the most part as 
modifications of characters held in common, rather than in the 
presence or absence of special characters. 
So it seems not improbable that, as the proper stages in de- 
velopment are reached, the secretions or emanations from the 
chromosomes of paternal and maternal origin can produce in 
combination or interaction with the cytoplasm and the general 
environment the ultimate characters peculiar to the respective 
parents. From what we already know of physiological chemistry 
we can but conclude that before the final adult organism is 
evolved there must ensue numerous reactions between the nuclear 
material and the cytoplasm, and probably many highly complex 
interactions would follow, both among the products of these re- 
actions and between them and other nuclear and cytoplasmic 
materials which had been developed in the meantime, or which 
had remained quiescent because of the want of proper material 
with which to react. 
By way of rough analogy, the writer has always thought of 
the process of specific organogeny as possibly somewhat similar 
to the production of galls on plants. Any one of several species 
of insects may produce galls on a given plant, but each kind 
of insect always produces its own specific type of gall. Here is an 
actual case of living protoplasm producing a specific character 
7 through the activity of a specific exciting agent. That is, the 
reaction between certain secretions of the insect and the living- 
substance of the plant produces new and definite structures. 
Change either factor and the resulting structure must be modified. 
Likewise, in the germ-cell, alterations in the constitution 
of either chromosomes or cytoplasm must undoubtedly pro- 
duce structural changes in the adult. It would seem that one 
might expect to find in the chromosomes the greater source of 
variability because they are derived in much greater proportions 
from two sources-the two parents-than is the cytoplasm, and 
this very mingling of the two materials could afford an adequate 
basis of both quantitative and qualitative changes. The change- 
able nature of chromosomes is evidenced, furthermore, by their 
marked differences in structure and appearance under different 
conditions of activity, and their instability is shown in the pro- 
nounced irregularities which may arise through hybridizing or 
from such external influence as drugging, etc. 
In so far as the cytoplasmic constitution is altered before 
the germ-cells are specifically set aside, we might reasonably 
expect this constitution to persist as such in the cytoplasm of 
these cells. We can readily see, moreover, if persistence of 
such a cytoplasmic constitution is possible, how changes of 
chromosomal constitution (or, indirectly, even the effects of 
physical environment) might be reflected onto the cytoplasm 
from time to time, and there conserved. 
A pursuit of this line of thought is tempting, but it would 
lead us far astray from our subject-matter. Nevertheless, it 
is well to remember that from the very fact that racial evolu- 
tion is possible we must concede an accumulative capacity of 
some kind to the germ-cell. -However our respective judgments 
may balance the factors of preformation or epigenesis in ontogeny, 
the fact confronts us that phylogenetic development has been 
largely an epigenesis, since in the first living matter there could 
not have been specific organ-forming substances for all the later 
organisms which have evolved from it. If organ-forming sub- 
stances are necessary for the ontogeny of present forms, it is a 
necessity which has been acquired hand in hand with the evolu- 
tion of the race, and the problem which confronts us is to solve 
the riddle of how it has been cumulatively grafted onto the 
pristine protoplasm. 
8 4. The presumed necessity of a reduction in the number of 
qualities which arise as the result of dual ancestry, together 
with the fact that there are reduction divisions strikingly 
similar in appearance and results in the germ-cells of both male 
and female, is also regarded as a proof that the chromosomes 
are the bearers of the hereditary characters. As a consequence 
of these reduction divisions, the male and female germ-cells each 
come to have only half the number of chromosomes character- 
istic of the ordinary tissue-cells, a fact which is interpreted by 
Weismann and his school as indicating that half the hereditary 
qualities have been eliminated in each germ-cell. The total 
number of chromosomes is restored again through the act of 
fertilization. 
Even ignoring the fact that reduction divisions of chro- 
mosomes do not occur in a number of seemingly well-authen- 
ticated cases, the theoretical necessity for qualitative reduction 
is by no means obvious. In the first place, it is improbable that 
the number of qualities are doubled at conjugation because, for 
the most part, leaving out of consideration superficial individual 
differences, probably like protoplasm is added to like protoplasm; 
or possibly, as already suggested, the male brings in materials 
which can participate in the construction of a more limited 
number of characters. As to the difficulty regarding the ulti- 
mate accumulation of individual qualities, there is evidence 
that they may often blend in time, or that certain ones become 
dominant. Finally, as far as probability is concerned, it is just 
as likely that inactive qualities, if represented materially, might 
be resorbed by the protoplasm (just as we have resorption of 
visible morphological structures, such as the tail of the tadpole) 
as that a complicated mechanism for throwing out half the quali- 
ties of each germ-cell is necessary. 
Weismann himself has shown, in fact, how, according to 
his theory of heredity, through panmixia (cessation of natural 
selection) and the battling of "determinants" for nutriment, a 
useless part "must grow smaller and smaller until finally it 
disappears altogether." {Germinal Selection-, second edition, p. 
42.) If one accepts this view of the elimination of superfluous 
qualities, it is difficult to see the necessity of postulating a second 
method of elimination by reduction divisions, or how, indeed, 
in the course of evolution the enormous supply of accessory or 
9 reserve characters which are attributed to the germ-plasm, 
could have arisen. 
5. The perpetuation through successive cell-generations 
of a fixed number of chromosomes, each apparently of distinct 
individuality, together with their final seeming accordance 
with the idea of "pure" germ-cells as demanded by Mendelian 
principles, is regarded by many as one of the strongest supports 
of the chromosome theory of heredity. 
That there is at least a kind of individuality (recurrence of 
form) of the chromosome is clearly demonstrated in the results 
of such investigators as Herla ('93), Zoja ('95), or Moenkhaus 
('04), who have studied the chromosomes of hybrids from widely 
separated forms which have noticeably different chromosomes. 
And the same thing is shown almost as conclusively, I think, 
in the researches of Montgomery ('01), Sutton ('02), Wilson 
('06), and others who, working upon non-hybrid forms which 
have chromosomes of varying size, have accumulated much 
evidence to the effect that in somatic and early germ-cells pairs 
of homologous chromosomes exist, one member of each pair 
being maternal, the other paternal, in origin. Persistency of 
form, however, does not necessarily imply persistency of consti- 
tution. This is evident when we consider that a chromosome is 
a complex made up of at least two substances, viz., an apparently 
more homogeneous linin substratum, encasing innumerable gran- 
ules of chromatin. Whether any exchange of individual gran- 
ules occurs, or whether any permanent changes are wrought in 
the chemical constitution of these granules or of the surrounding 
cytoplasm, we have at present no direct means of ascertaining. 
It is certain that abundant opportunity for such alterations is 
afforded when chromosomes are resolved into their ordinary 
diffuse condition in the nucleus. 
Granted that in hybrid offspring there are such things as 
germ-cells "pure" with respect to a given character-and there 
are some who would dispute this-it would seem, if we assume 
that the chromosomes determine certain hereditary characters, 
that in the separation of homologous chromosomes of different 
parentage an adequate mechanism exists for the segregation of 
these qualities, as has recently been advocated by various 
cytologists. Doubt is cast upon this interpretation, however, 
10 by such facts of non-Mendelian inheritance as blending, the 
persistence of hybrid mosaics, and the lingering of certain 
influences in gametes which theoretically should have been purged 
of such characters. 
The very fact, however, that constant types of visibly 
different chromosomes recur time after time in cell divisions 
shows, at least, that under certain conditions they possess differ- 
ent physical properties, and that in some respects, therefore, they 
are qualitatively different. But there are no sufficient reasons, 
I think, why yve may not look upon their differences as differ- 
ences of more elemental chemical and physical constitution 
rather than as differences among systems of determinate morpho- 
logical units. Because when the two are brought together under 
certain conditions water (H2O) is the result, we do not postu- 
late a "determinant" of "aquosity" in hydrogen or in oxygen. 
It is true that in view of certain properties possessed by each of 
the two gases, the formation of water is possible, but bring them 
together under other conditions and the same two elements yield 
an entirely different "character," viz., hydrogen peroxide (H2O2). 
Or, again, let chlorine act upon benzene (CfiH6) and, depending 
upon purely quantitative relations and other physical conditions, 
any one of six different substitution products ranging from 
C6H3C1 to C6C16 can be secured. No quasi-teleological con- 
ception of anticipatory units is found necessary in a considera- 
tion of such cases of chemical configuration, and it seems to me 
that its necessity, as implied in the conception of determinants 
or pangenes, is yet to be demonstrated in phenomena of heredity. 
Even in case of the divorcement of paired parental chromosomes 
in gametes-and I think this is strongly evidenced in a number 
of Tracheata-it would seem that we might account for the so- 
called Mendelian phenomena by attributing to the chromosomes 
simply chemical and physical differences without endowing them 
with morphological entities. And, moreover, on such a basis, 
looking to the reactions of nucleus and cytoplasmic materials 
for the establishment of ultimate "characters," we can more 
readily see how non-Mendelian gradations and "contaminations" 
of characters might arise, because we might then attribute more 
importance to purely quantitative relations and chance mixture 
of the chemical substances. In short, there can be a "quali- 
tative" basis, or a series of qualitative bases, without these 
11 being at the same time specific "determinants," and until such 
a more neutral qualitative basis for the phenomena of heredity 
is shown to be untenable there is no valid reason for postulating 
a series of morphogenic entities. 
But such considerations as these open up another important 
question, viz., as to just how we are to regard heritable qualities. 
One great difficulty here lies in the vagueness which enshrouds 
the term "quality" or "character." It is obvious that many 
so-called characters are in reality only the expression of the re- 
lations of a number of parts and can have no individual basis 
of their own, 
In all our systematic zoology stability is the fundamental 
principle used in selecting characters for purposes of classifica- 
tion. For example, the features chosen as characteristic of 
genera have shown themselves to be more constant than those 
which are selected to indicate species, otherwise they would have 
been discarded as generic characters; specific characters, in turn, 
are less fluctuating than varietal traits. The qualities which 
stamp a given animal as a vertebrate or a mammal are cer- 
tainly more stable and definitely coordinated than those which 
mark it as a particular variety or species. Our whole scheme of 
natural classification, which is, when correct, but an expression 
of the evolutionary status of the forms classified, is based neces- 
sarily upon the facts of heredity-that is, of community of descent. 
Does not the very fact itself that certain character relations are 
uniform and others fluctuating through successive generations 
show that in seeking for a physical-unit basis of inheritance we 
cannot expect to find it wholly in a series of equipotent units, 
but that unmistakably we have to deal with groupings of char- 
acters, some groups of which are more stable than others? This, 
in turn, can mean only that any such group must be a unit in 
itself, and that we are dealing, therefore, with units of a higher 
and of a lower order; or, in other words, with series of coordina- 
tions built upon broader coordinations. 
To some, such an admission may seem inevitably to demand 
as a consequence some kind of morphological basis or organiza- 
tion in the germ-cell, some coordinating influence, to which the 
more restricted chemical processes are subordinated. However 
this may be, it assuredly does not make an accompanying de- 
mand that such controlling factor or factors shall reside wholly 
12 or predominantly in the chromosomes. The so-called "organ- 
ism" standpoint, indeed, which as new facts come to light is 
appealing to more and more minds, would seem to tend rather 
toward conclusions just the reverse of this or of any that would 
seek to localize the adult morphology of a living organism in 
any special part, of its antecedent germ-cell. A germ-cell, in 
fact, should need no special units to generate the peculiar genre 
equilibrium or idiosyncrasy of protoplasm which is distinctive 
of a particular kind of individual, since such a germ-cell not only 
is itself already an individual, but from the very fact of having 
had the same racial history as other individuals of its peculiar 
kind (be they germ-cell, embryo, or adult) it must likewise as 
a whole already possess this distinctive idiosyncracy. 
The results of recent experiments on regeneration and regula- 
tion tend, in so far as they have bearing on ultimate organogeny, 
to emphasize the significance of an organism as a whole to its 
environment, together with its physiological coordination or 
interaction of parts. The results of such work seem to indicate 
clearly, as ably maintained by Child ('06), a fundamental physi- 
ological unity of the entire organism to account for which any 
purely "unit-character" basis of transmission is inadequate. 
To express the phenomena of organic characters, it would 
seem that we must turn to a condition somewhat analogous to 
that with which we meet in chemistry in many organic compounds; 
and while the comparison is only an analogy, we might well 
remember that it is an analogy drawn from organic matter it- 
self. For example, in many organic compounds we have cer- 
tain fundamental groups of comparatively great stability, well 
illustrated in the so-called benzene ring. We can substitute for 
the hydrogens of this ring one, two, or many alkyl or other 
groups, thus producing different compounds. Yet these com- 
pounds have certain fundamental properties in common, due to 
the benzene ring. 
While in living beings qualities must be looked upon as 
more or less flexible rather than absolutely static, with certain 
possibilities of blending, of gradational and of cumulative effects, 
still we must recognize more and less stable coordinations. 
But even in this matter of flexibility chemical and physical 
analogy does not forsake us, for we find many analogous examples 
of similar flexible or gradational combinations, in which the 
13 components may establish any proportional relation one to 
another, depending upon external factors. This is best evi- 
denced in solutions, and especially in certain solid solutions. 
Moreover, in phenomena of "dynamic isomerism,'' in changing 
substances of the same chemical composition but of different 
structure one into another, we can get a condition of equili- 
brium between the two at any point, the condition of this equili- 
brium depending upon temperature, pressure, and concentration. 
Again, in allotropic forms of such substances as sulphur, phos- 
phorus, tin, etc., various conditions of stability can be brought 
about in different ways. 
6. Boveri's demonstration ('89, '95), that the denuclc- 
ated egg of one species of sea urchin when fertilized with the 
spermatozoon of another species shows purely paternal char- 
acters, has been numbered among the proofs of the exclusive 
control by the nucleus of matters of heredity. 
Both Seeliger ('95, '96) and Morgan ('95), however, have 
shown that even when the egg is not denucleated, still, in hybrids 
between the forms with which Boveri worked, the paternal type 
may occasionally predominate, to the apparent exclusion of the 
maternal type; that, in fact, Boveri's experiments may mean 
simply that in such cross fertilization the characters of the male 
species are prepotent over those of the female species. Fur- 
thermore, as just the converse of Boveri's results, we have 
Godlewski's ('05) observations that non-nucleated pieces of 
sea-urchin eggs, fertilized by sperm from a crinoid, produced 
larvae exclusively of the maternal type. 
7. The apparent association of specific qualities of the 
adult with specific chromosomes is regarded as another support 
of the chromosome theory of heredity. The most noted ex- 
ample of this is the finding of a chromosome that is regarded as 
a possible sex determinant in certain tracheata. 
We cannot enter here into a review of this intricate matter, 
a detailed discussion of which will be found in Professor Wilson's 
('06) paper. While the whole question of whether sex is pre- 
determined, or whether it may be determined after development 
has begun, remains almost as much of a puzzle as ever; still, 
if there is any truth in the latter alternative, the trend of ex- 
14 periment points mainly to the conclusion that nutrition is the 
chief factor in determining sex. Consequently, since of the in- 
sects exhibiting the "accessory chromosome" the females are 
characterized by the presence of an extra chromosome (or by 
a greater bulk of active chromatin in case idiochromosomes are 
present), we seem justified in asking that it be shown untenable 
that the production of females is due, not necessarily to special 
sex-determinants in the chromosome, but to the presence of 
more chromatin which has meant increased chemical activity 
on the part of the nucleus. In the few cases where the idio- 
chromosomes are of equal size, however, purely quantitative re- 
lations are apparently inadequate as an explanation. But even 
here there may be a difference in intensity of chemical activity 
between the two chromosomes, since, in this group of insects, 
according to Wilson, there are indications of a tendency of one 
of the idiochromosomes to disappear ultimately. 
The observations of Boveri ('02) on echinoid eggs which 
have been fertilized by two spermatozoa is also regarded by 
some as strong evidence that the nucleus is the real bearer of 
hereditary qualities. We must recognize, however, that in such 
cases as the production of three cells by means of a tri-polar 
spindle, for example, the cytoplasm of the sperm, scant as it 
may be, is also distributed among the three cells, as is also the 
substance of the spindles; so that the result is not three different 
series and combinations of chromosomes in three cells of similar 
protoplasm, but a series of chromosomes in cytoplasm, which 
itself differs in the three cells. Again, if the individual chromo- 
somes differ only in that they produce different nutritive mate- 
rials, or enzymes, then we might expect different and abnormal 
results from different blastomeres, since the cytoplasmic mechan- 
ism could only react on or be stimulated by the substances dis- 
tributed to it. Thus absence of parts in a portion of the body 
developed from a blastomere in which an insufficient number 
of chromosomes is present might well be expected, and the whole 
phenomenon is as easily interpreted to mean that, considering 
nucleus and cytoplasm as of coordinate importance in inheri- 
tance, a proper reaction has been prevented or that insufficient 
material was present, as to regard it as a demonstration of the 
exclusive importance of the nucleus in heredity. 
Lastly, inconstancy in the number of chromosomes in closely 
15 allied forms argues against the idea that we shall ultimately be 
able to associate specific characters of the adult with individual 
chromosomes. The numerical differences would seem to be out 
of all proportion to the actual differences between the adults 
of the species or genera showing such discrepancies. 
On the other hand, if we find closely allied genera or species 
with chromosomes very constant in number and appearance, 
even should they have no causal connection with the phenomena 
of heredity, one need be no more surprised than at finding close 
similarity among any other organs in closely related species. 
It might be argued, indeed, with some plausibility, that the 
number and arrangement of the chromosomes in a given species 
are the effects of the fundamental constitution of a given kind 
of living matter, rather than that they stand in a specifically 
casual relation to such constitution. 
By way of final summary with regard to the proposition 
that the chromosomes are the exclusive vehicles of heredity, we 
may, I think, deny that a satisfactory case has as yet been proven. 
While the jumble of facts which have been determined so far 
may not negate the theory, still we are not justified for that 
reason in maintaining that they substantiate it, especially when 
most of the facts are so Janus-faced as to be of equal utility in 
confirming other hypotheses. 
The important fact always confronts us that a given kind 
of protoplasm is a protoplasm peculiar to the organism of which 
it forms a part, whether the latter be amoeba or man, and the 
egg as a whole, both cytoplasm and nucleus, therefore, has its 
own individuality. Hence no special formative force has to 
change it into a specific kind of protoplasm before the more ob- 
vious morphological entities can become manifest. Heredity is a 
problem of the handing on of metabolic energies already estab- 
lished, rather than of the transmission of a series of determinative 
units which create a wholly new organism. We can see that in 
so far as the substances constructed by the nucleus are peculiar 
or individual, the number of structures the cytoplasm can shape 
from such material or form in combination with it has been re- 
stricted, and in this sense the nucleus has conditioned heredity. 
But because the three elements, carbon, oxygen, and hydrogen, 
condition substances of which they are components, we do not 
16 postulate a specifically determinative substance in any of them 
for each of the numerous carbohydrates and carbohydrate pro- 
ducts that result from their various combinations and arrange- 
ments. What the chemist seeks to determine are the quanti- 
tative and other physical relations necessary to the establish- 
ment of a certain molecular configuration. Likewise, what we 
seek in heredity are the shaping and controlling factors which 
bring materials into place that they may combine in the proper 
way and at the proper time. This much is certain: no chemical, 
physiological, or morphological evidence is yet extant which 
places these factors wholly within the chromosomes. 
I do not desire to minimize the importance of the chromo- 
somes in heredity, and I think no one would deny that they may 
stand in definite causal relationship to adult characters. On 
the other hand, I see no sufficient reasons for denying that other 
germ-cell constituents may, in the same sense, stand in causal 
relationship to such characters. In these initial substances, 
however, I see no more necessity for postulating specific anti- 
cipatory characters (beyond the properties which makes a given 
substance a substance per se, irrespective of the structures into 
which it may ultimately be builded) than I do of regarding 
yeast, or flour, or milk as in itself a specific determinant of a 
loaf of bread. While the net results may be in large measure 
the same, whether we accept the rigid "determinant" idea or 
whether we adhere to a more neutral qualitative basis, there 
are certain elements of freedom in the latter conception, I think, 
which render it a safer foundation for unbiased investigation of 
the problems of heredity. I would plead, therefore, not for the 
abandonment but for the maintainance of other working hypothe- 
ses as our greatest safeguard. 
17 LITERATURE CITED 
Auerbach, L., '96. Untersuchungen uber die Spermatogenesc 
von Paludina. Jen. Zeits. xxx. 
Boveri, Th., '89. Ein geschlechtlich erzeugter Organismus 
ohne mutterliche Eigenschaften. Sitz. Ber. Ges. Morph. 
Phys., Munchen, v. 
Boveri, Th., '95. Uber die Befruchtungs - und Entwickcl- 
ungsfahigkeit kernloser Seeigcl-Eier, etc. Arch. Entwni. 
Boveri, Th., '02. Uber mehrpolige Mitosis als Mittel zur 
Analyse des Zellkerns. Vehr. der. Phys. Med. Ges. zu 
Wurzburg, N. F., Bd. xxxv. 
Child, C. M., '06. The Relation between Functional Regu- 
lation and Form-regulation. Journ. Exp. Zool. in, 4. 
Child, C. M., '06. Contribution toward a Theory of Regu- 
lation. Arch, fur Entw.-Mech. xx, 3. 
Child, C. M., '07. Studies on the Relation between Amitosis 
and Mitosis. Biol. Bui. xu, 2, 3. 
Conklin, E. G., '05. The Organization and Cell-lineage of 
the Ascidia)i Egg. Journ. Acad. Nat. Sciences of Phil., 
Second Series, xm, 1. 
Davenport, C. B.,'06. Inheritance in Poultry. Carnegie Insti- 
tution of Washing-ton, Washington, D. C. 
Godlewski, E. Jun., '06. Die Hybridisation der Echiniden 
und Crinoiden Familie. Arch, fur Entw.-Mech. xx. 
Guyer, M. F., '07. Do Offspring Inherit Equally from each 
Parent? Science, N. S., Vol. xxv, No. 652. 
Herla, V., '93. Etude des variations de la mitose chez VAsca- 
ride megalo cep hale. Arch. d. Biol, xm, pp. 423-520. 
Lillie, Frank R., '06. Observations and Experiments Con- 
cerning the Elementary Phenomena of Embryonic Devel- 
opment in Chaetopterus. Journ. Exp. Zool. m, 2. 
18 Loeb, J., '99. Warum 1st die Regeneration kernloser Proto 
plasmastucke unmoglich oder erschzuert? Arch. Ent. 
Organ, vm, 4. 
Meves, F., '94. Uber eine Metamorphose der Attraktions- 
sphdre in den Spermatogonien von Salamandra macu- 
losa. Arch, fur mikr. Anat., Bd. 44. 
Moenkhaus, William J.,'04. The Development of the Hybrids 
between Fundulus heteroclitus and Menidia notata, zvith 
Especial Reference to the Behavior of the Maternal ana 
Paternal Chromatin. Am. Journ. Anat, m, 1. 
Montgomery, T. H., '01. A Study of the Germ Cells oj 
Metazoa. Trans. Am. Phil. Soc. xx. 
Morgan, T. H., '95. The Fertilization of Non-nucleated Frag- 
ments of Echinoderm Eggs. Arch, fiir Entw.-Mech. n. 
Roux, Wilhelm, '83. Uber die Bedeutung der Kerntheilungs- 
figuren. Leipzig. 
Seeliger, O., '95. Gibt es geschlechtlich erzeugte Organis- 
men ohne mutterliche Eigenschaften? Arch, fiir Entw.- 
Mech. i. 
Seeliger, O., '96. Bemerkung uber Bastardlarven der See- 
igel. Arch, fiir Entw.-Mech. in. 
Spitzer, -, '97. Pfliiger's Archiv., Vol. 67, p. 615. 
Sutton, W. S., '02. On the Morphology of the Chromosomc- 
group in Brachystola magna. Biol. Bui. iv, 1. 
Weismann, August, -. On Germinal Selections. Open Court 
Pub. Co., Chicago, Ill. 
Wilson, E. B., '06. Studies on Chromosomes, in. The Sexual 
Differences of the Chromosome-groups in Hemiptera, with 
some Considerations on the Determination and Inheritance 
of Sex. Journ. Exp. Zool. in, 1. 
Zoga, R., '95. Sulla independenza della cromatina paterna e 
materna nel nucleo delle cellule embryonale. Anat. Anz. 
il, pp. 289-293. 
19