FERRuUARY 4, 1904] 
NATURE 
Bait 
But in spite of these irregularities that tend to 
obscure the more important facts, the heterotype 
division can be recognised with certainty in every 
malignant growth so far examined; and it is pre- 
cisely similar in character to the normal heterotype that 
occurs in the sexually reproductive cell series. The 
same peculiarities in the early differentiation of the 
chromosomes culminating in the production of rings, 
loops, &c., the same reduction in the number, and the 
same transverse division of each one when attached to 
the spindle, reappears in these cells with the greatest | 
uniformity. 
This peculiar mitosis seems to be confined in tumours 
to those of a malignant character, for it has not been 
observed up to the present in any benign growth. It 
would thus appear to serve as a means of distinguish- 
ing between the two classes of growths. 
Following upon the heterotype division, the homo- 
type stage is reached, but it very soon becomes unrecog- 
nisable in most cases owing to the occurrence of the 
irregularities above mentioned. 
The conclusion to be drawn from the above account is 
that, in a most important respect, some of the cells of a 
malignant growth have gone through a change similar 
to that which in normal tissues is confined to the pro- 
duction of the generations ending with the formation of 
the sexual cells. Such a conclusion is further supported 
by considerations derived from other sources. 
It has already been pointed out that whereas in 
animals the differentiation of the sexual elements fol- 
lows closely after the occurrence of the heterotype mi- 
tosis, this is not the case in most plants. Thus ina 
fern, the whole prothallium is composed of post-hetero- 
type cells, and the sexual elements only arise from a 
relatively small number of them. Similarly in the 
embryosac of a flowering plant, there are certain post- 
heterotype cells that are not normally destined to give 
rise to sexual structures. But it is a matter of con- 
siderable interest to find that cells that fail in this 
respect not seldom exhibit marked irregularities in their 
modes of further division. Sometimes direct fission of 
the nuclei may occur with suppression of chromosome 
differentiation ; in other cases the chromosomes may 
appear, but in quite irregular numbers. 
The similarity of these irregularities to those 
already indicated as present in cancerous growths will 
at once be obvious from what has already been said. 
The investigations of Bashford and Murray have 
served to confirm the statements previously made as to 
the occurrence of heterotype and homotype mitoses in 
the human subject. These investigators have iden- 
tified the same divisions in malignant growths that 
occur in other mammals, in reptiles, and in fish. 
Whether, therefore, the explanation advanced to ex- 
plain them, which involves the admission of an essen- 
tial similarity as existing between the malignant 
growths and sexual reproductive tissue, be accepted or 
not, it is a fact that will have to be reckoned with. 
It has been held by some persons that a transform- 
ation of somatic into reproductive tissues cannot occur, 
and it is, therefore, necessary to examine briefly the 
grounds on which such an opinion rests. : 
In plants the difficulty does not really arise, for a 
large number of cases are known in which cells that 
have long discharged somatic functions may revert to 
an embryonic condition, and then, after a heterotype 
division, produce from amongst their descendants the 
sexual elements that take part in fertilisation. | This 
fact robs the objection of any a priori force it might have 
had. It is, however, true that amongst animals the 
conversion does not normally occur, but the existence 
of the diagnostic mitosis described above as appearing | 
in the malignant growths affords cogent evidence for 
No. 1788, voL. 69] 
regarding them as representing such a changed condi- 
tion, the true nature of which is, however, masked by 
the invariably pathological features that accompany 
it. 
It is not urged that the cancer cells are functionally 
active sexual elements, but rather that they are homo- 
logous with such; it has, therefore, been proposed to 
express this idea by applying the term ‘‘ gametoid ’” 
to them. 
But whilst the existence of the heterotype mitosis em- 
phasises the gametoid nature of the cells that have just 
passed through it, there are other phenomena that sug- 
gest the interpretation may possibly be carried on to 
another and further stage. Just as the true gametes 
| (sexual cells) may fuse, so, too, cases of nuclear fusion 
are not very uncommon in the post-heterotype cells of 
malignant growths. It would be premature at the pre- 
sent juncture to attempt to do more than indicate that 
there may be something beyond a mere abnormality 
latent in these fusions. It is, however, a fact that in 
individual cases the fusion figures strongly recall in- 
stances of normal fertilisation. Should the suggestion 
turn out to be well founded, and many instances 
| apparently support it, much that is still difficult of ex- 
planation will immediately become clear. The irregular 
nuclear divisions, for example, will be no more surpris- 
ing than are those so frequently to be seen in the endo- 
sperm of an angiosperm, or even in the more abnor- 
mal results consequent on polyspermy. The inde- 
pendence of the neoplasm and its parasitic habit, to 
which attention has already been directed, would be 
still quite explicable, for in a general sense it may be 
stated that a new generation habitually preys on its 
forbears whenever continued association with them ad- 
mits of it. 
But the problems that especially invite attack 
are those concerned with the causes of the trans- 
formation of somatic, into reproductive, cells and 
tissues. These fall within the scope of the physiological 
chemistry of the cell. Something has already been 
done in this direction so far as plants are concerned ; 
and, indeed, it would seem that the lower members of 
the vegetable kingdom offer a more convenient material 
for investigation than animals. They are compara- 
tively easy subjects of experiment, and their simpler 
specialisation avoids the difficulties consequent on the 
presence of complicated subsidiary mechanisms. The 
ease with which Spirogyra, for example, can be directed 
into either the reproductive or the vegetative phase is a 
case in point, and it is only one out of many that could 
be cited. J. B. Farmer. 
SCIENCE AND MILITARY EDUCATION. 
Jpele Journal of the Royal United Service Institution 
for January contains a full account of the impor- 
tant discussion on November 9, 1903, initiated by 
Lieut.-Colonel F. N. Maude, late R.E., on, the subject 
of military education, and on January 18 there was 
published a revised scheme of subjects for the entrance 
examinations to the Royal Military Academy and the 
Royal Military College respectively. _ The discussion 
at the United Service Institution, which was of a 
decidedly discursive character, dealt to a large extent 
with a real or supposed deterioration of the public 
school boy of to-day, or at least of those public school 
boys who desire to obtain commissions in His Majesty’s 
Army. 
This part of the discussion was based very largely 
on statements made by army tutors, which, though 
there may be some truth in them, must be rather care- 
fully scrutinised. First, because army tutors are human, 
