strosities can hardly be considered as the result of an arrest
of development; for parts of which no trace can be detected in the embryo,
but which occur in other members of the same class of animals or plants,
occasionally appear, and these may probably with truth be attributed to
reversion. For instance: supernumerary mammae, capable of secreting milk,
are not extremely rare in women; and as many as five have been observed.
When four are developed, they are generally arranged symmetrically on each
side of the chest; and in one instance a woman (the daughter of another
with supernumerary mammae) had one mamma, which yielded milk, developed in
the inguinal region. This latter case, when we remember the position of the
mammae in some of the lower animals on both the chest and inguinal region,
is highly remarkable, and leads to the belief that in all cases the
additional mammae in woman are due to reversion. The facts given in the
last chapter on the tendency in supernumerary digits to regrowth after
amputation, indicate their relation to the digits of the lower vertebrate
animals, and lead to the suspicion that their appearance may in some manner
be connected with reversion. But I shall have to recur, in the chapter on
pangenesis, to the abnormal multiplication of organs, and likewise to their
occasional transposition. The occasional development in man of the
coccygeal vertebrae into a short and free tail, though it thus becomes in
one sense more perfectly developed, may at the same time be considered as
an arrest of development, and as a case of reversion. The greater frequency
of a monstrous kind of proboscis in the pig than in any other mammal,
considering the position of the pig {58} in the mammalian series, has
likewise been attributed, perhaps truly, to reversion.[130]

When flowers which are properly irregular in structure become regular
or peloric, the change is generally looked at by botanists as a return
to the primitive state. But Dr. Maxwell Masters,[131] who has ably
discussed this subject, remarks that when, for instance, all the sepals
of a Tropaeolum become green and of the same shape, instead of being
coloured with one alone prolonged into a spur, or when all the petals
of a Linaria become simple and regular, such cases may be due merely to
an arrest of development; for in these flowers all the organs during
their earliest condition are symmetrical, and, if arrested at th