iS>b6erv>ation6

...on tbc...

dolors of

3f lowers anb 3Leav>ee

E. WilUams Merve'e.

OBSERWATIONS

COLORS OF FLOWERS

E. WILLIAMS HERUEY.

NEW BEDFORD:

K. Anthonv & Sons, Incorp., Pkintkks.
1899.

Copyriglit, 1899, by E. WILLIAMS Hervev

INTRODUCTION.

Color in Nature, and especially in the vegetable world
as seen in leaf and flower, is one of the most pleasing-
objects that greet the eye.

Many questions naturally arise regarding its origin and
the purposes it subserves, its modifications and many
changes. These offer a somewhat novel subject for en-
quiry. There is very little literature on the subject, and
a portion of what exists is open to grave criticism.

Chemists have informed us of the nature of only a few
of the pigments and where they may be found, whether
fixed in minute bodies called plastids, or free in the cell
sap.

Grant Allen has written very entertainingly on the se-
quence of color and Hermann Miiller has discoursed on
insect selection ; the writer however, not being wholh-
convinced as to the soundness of the theories of these
authorities, recalling moreover Agassiz's injunction, which
he copied from a blackboard in the Penikese school :
"Study to translate what actually exists," "Study nature,
not books," decided to investigate for himself. Herein he
presents the results of his researches, including a general
survey of color in flowers, leaves, etc., as seen by an ordi-
nary observer, reviewing also some of the more obvious
facts and comparing them with the conclusions of the above
mentioned and other writers. Some of the ways in which
flowers become variegated will also be pointed out, and in
brief there will be presented a general conspectus of color
astound in vegetation.

Elipiialet Williams Hervey.

New Bedford, Mass., May 1, 1899.

OBSERVATIONS

ON

THE COLORS OF FLOWERS.

SEQUENCE OF COLORS.

CHAPTEK I,

It has l)een thought that as tlowers arc morphoktgically
modified leaves, that green is the primitive color; others
favor yellow, making purple and blue later developments.

Let us first consider the relation of green to other colors.
It is the most prominent of all hues ; we find that it is not
confined to foliage, but that a very large and important
portion of the Floral Kingdom existing at the present day
is wholly green,- or has green as its ground color, and that
this color is not restricted to small, weed-like plants which
may be degraded specimens of more highly developed an-
cestors, but it is also found among the choicest plants of
gai-dens and conservatories.

Neither are these green fiowers limited to primary forms,
but they are found in every stage of advancement from the
simplest to the most complex. The green color of the bud
of course, as every one knows, as a rule precedes all other
colors, and when in this green state it is generally impos-
sible to say what color will follow, for however bright the
future color may prove to be, for the time being, it is com-
pletely concealed, but greeri in its more permanent forms
also yields directly to any hue.

We will glance for a moment at a few prominent green
fiowers. Some of the most conspicuous of these found in

our conservatories are the Cypripediums ; there are at least
a dozen s|)ecies having as pure a green color as the general
foliage of plants, except that one or more petals may be
relieved by a slight blending with another tint which Nature
adds to lessen their homeliness, and at the same time to
make them more attractive to the insects which visit them
and assist in their fertilization.
C. longifolium is wholly green.

G. insigne: the green dorsal sepal which stands cons[)ic-
uously erect is tipped for a third of its length with
pure white.
C Harrissianum : the veins of the dorsal sepal, which is
largely white, and also of the sli})per or tubular lower
petal, are turned to a dull purple.
C. Lawrenceamim : the dorsal sei)al has become quite
white l)y the fading out of the green, except that
there remains a few streaks with the purple veins.
There is moreover a variety of C caUosiwi in which the
whole tlower is of the purest white, saving the green veins.
C. Charles worthii: the whole dorsal petal has changed

from white to a beautiful i)uik purple.
C. Sedenii and C. Calurum have all the petals of a rose

color, only a trace of green remaining.
The green portions of these tlowers by age become yel-
lowish, for orchids as a rule remain fresh for a month or
six weeks. Green flowers often become yellowish-green or
greenish-yellow and finally pure yellow ; and among the
Cypripediums are found [)ure yellow varieties.

One can therefore trace from the foundation green, a
white, a purple, a rose, and a yellow, and an iiitinite num-
ber of shades and tints. The purples and reds are evi-
dently separate and independent colors, breaking through
the chlorophyl green.

Yellow often results from the disintegration or dissolution
of the chlorophyl itself, which is largely composed of
yellow with some I)lue. White results from the loss of

color, as seen when either yellow, green, red, purple, or
blue or other color fades and disappears from the petal.
This white may be pigmental or the white of the original
color of the tissue, the chlorotic white, as seen in the em-
bryo of the seed, and commonly in the earliest growth of
roots and stems. A common garden bean when split open
will show not only the thick white cotyledons, l)ut a st'cond
pair of delicately veined leaves white as marble, which on
germinating change to green. One-half of all so-called
white flowers are no whiter than this primitive white tissue.

There are differently constituted whites as well as yellows.
This last mentioned white color is not produced by a ])ig-
ment ; there is no coloring matter, but the whiteness is an
optical effect similar to that seen in snow which dissolves
into colorless water. It is caused by minute bubbles of air
in the (!ell walls which reflect white light. There does not
api)ear, however, to be any dividing line between a trans-
lucent, structural white, as seen in the embryo, the Easter
lily, etc., and a i)igmental chalky white as seen in the rays
of a daisy, which latter is occasioned by numerous small
granular l)odies in the cells of the tissue known as louco-
phytes.

Lest the exam})les of Cyprij)e(liums might appoar to be
exceptional, we add a list of other green ilowers l)oth showy
and inconspicuous.

Zy(iopetalum, green with darker markings of brown

Lyamie gigantea. olive-green, lij) maroon ;ind orange.

Do. ciliata, green, li}) white and buff yellow.

Ci/cnodie, four species, green suffused with purple.

Angroicum sesquq)edale, light green, spur ten inchi's long.

Gtdeandra, several s})ecies, green with lips white or rose.

Caitlei/a bicoloi- cwndea, green, lip sky-gray.

Coilogi/ne pandiDutd, pale green, li[) yellow-green.

Eitcomis, eight or more species, green.

>Sel€)iij)ediu)its, ten species, green shaded with reddish-
brown.

Epidendriuii Pseud. Ejiidendrmn, apple-green, lip orange-
red.

Catasetum, Orchis, (Jlrrhopetalum, Cyndjidium, Pkity-
dinus, Masdevalia, Oncidkwi, Brassia, etc., all have
green species.

The following sections will be recognized as native plants :

Gymnocladus Canadensis, two shades of green, Hower has
a delicious spicy odor.

Xanthoxylum, Ptelea, Rhus, Vitis, Auqyelojjsis, Euony-
mus, Scleranthus, Sniilax, Chrysosplenium, Habenaria,
Physalis, Arismma, numerous species of Chenopodiacece,
Ambrosia, Amarantacem, Salicornia, Ludwir/ia, Ribes,
Helonias, purple changing to green, and Cobea scandens of
gardens green changing to purple, Liriodemlron Tidipifera,
etc., etc.

GKEENISH-PURPLE .

ScropJiularia, Menziesia, Aristolochia, Ap>Iec(rnm, Tril-
lium sessile; Veralrum, green turning brownish, purple
within, Tipndaria, Clematis, etc.

GREENISH-WIIITE.

Hydrastis, Polyijonatum, SyncJirys, Conobea, Pyrola,
Anemone, Comandra, T'riglodiin, (roodyera, Celastrus,
Smilacina, Echinocystis, Polygonum, Aralia, Solea, Poly-
gala, Iva ; Orniihogalum, green and white, GalanUius, sev-
eral species of a green color ; G. Elwesi is white marked
with green, G. virescens, outer petals pale green shading-
down to pure white, inner petals entirely green. Habenaria
lacera is pale green, except that the lip fades to white.

If green be the earliest color of tlowers, one would natu-
rally expect to find that color among the simplest forms, as
in Ranuncidacew, and it is found that many are so, as five
species or more of Clematis are green or greenish. Anemone
cylindrica greenish-white, A. Virginiana greenish, Thxdie-

trum jnirpvrascens jrroenish and purplish, Myosurus green-
ish, TrolHus greenish-yellow, Ilellehorus, several speeies,
greenish or wholly green, /i7io?r?^owi« greenish-yellow, Isop-
yrum greenish-white, Ilydmstis greenish-white, ete.

This change from green to white is common on the mar-
gins of green sepals, as in ]Sfymph<va odorata, Slellaria,
Buda, Phintago, etc.

Polygala sanguinea has three sepals green, the original
color, and the two lateral sepals purple. P. poly gaum has
three sepals green with rose-purple edges, the other two
Ijeing rose-purple like the petals. Dutchman's Pipe has the
curious green corolhi shaded and dotted with purple.

The large white involucre of Cornus Florida is like most
real flowers, green before becoming white. Indeed, in
Clematis several species described as purplish are lacking
the clear color because they are dulled by green !

In Pwonia Browtiii, of same family, the brownish-red of
the [)etals doubtless results from the mixing of chlorophyl-
green and red, an<l the l)rown-purple of Xaiifliorriza from
the mingling of green and purple.

We find positive evidence, therefore, that some of the
earliest and simplest members of Ranuncidacea' are still
green, and that others have green as a foundation color,
from which white, purple, red and yellow may in any order
originate ,

GREENISTI-VELLOW,

That a green flower may become yellow in a manner
analogous to a green leaf becoming yellow, is too well
known to require connnent. in some instances the yellow
is apparently a direct product of a previous green, resulting,
as in foliage leaves, from the disintegration of the chloro-
phyl and the disappearance of the blue element. This
yellow is reg:irded by some chemists as an entirely distinct
pigment and has been n;imed Xanthophyl. Of course the
yellow developed directly from white in the case of Loni-
cera longijiora, nmst be of a different character.

10

The tiansition from green to yellow may be traced in all
stages. A short list only is added of native plants : Pro-
sarles. Uvular ia, Dioscorea, Agave, MedeoJa, Sassafras,
Clmlonia, Physalis, Nicotiana, Frasera, Sedum, Penthormn,
Gonolobus, Shepherdia, Trollins, Myosurus, Liriodendron,
etc.

The leaves of many healthy plants turn from green to
yellow, as seen in numerous garden shrubs and vines which
are wholly yellow or variegated in various ways, as a green
leaf with a yellow margin. Fruits turn from green to
yellow, as the banana, lemon, etc.

Autumn leaves are also largely yellow, and most flowers
of a yellow color are preceded by green, as in JVup/tar,
Hyjwxis, Narcissus, lilies, etc., but there appears to be a
yellow of a different quality not derived from green, as
seen in roots, bulbs, early shoots, and when a white flower,
as Lonicera Jongifora, changes to a clear yellow.

GREEN, YELLOW AND BROWN.

We have already alluded to the brown and brown-purples
of some of our native plants ; but among tropical plants,
and especially the orchids, one is surprised at the number
of flowers having these colors. Very often the brown is
associate<l with yellow ; at times the flowers appear to have
yellow as the ground color, and to be merely blotched or
barred with brown, while in other species the flowers may
be wholly brown or brown with yellow markings. A few
descriptions will illustrate these combinations of color :

Ejndendrum oncidivides, red-brown marked with 3'ellow.

E. ocltraceum, orange-brown, lip yellow.

Phalmnopsis Luddemannia, chestnut-brown, with pale
yellow transverse streaks.

Oncidmm. Baiemanianmn, light yellow, blotched with
chestnut-brown.

O. Baneri, yellowish-green, spotted with red-brown.

O. harJ)atmn, yellow, blotched with chestnut-l)rown.

11

This genus has upwards of throe hundred species, nine-
tenths of which arc yelU)W and brown.

Vanda tricolor, light yellow, spotted with red-I)rf)wn.

Millonia cwieata, brown, tipped with pale yellow.

MasdevaUia, PhaiKs, Coelogyne, Calanthe, Schombvrgkia,
Zygopetalum, Paphinia, Catai^etum, Trrchocentrmn, Den-
drohium, Lycaste, Houllelia, Darlingtonia, etc., have more
or less species of these colors, to which should be added
Odontoglonswin, with upwards of sixty species and varieties
of brown and yellow.

If there be a law of progression in colors, as has been
maintained by one writer, where ought the various shades
of brown to be placed? Should they precede or follow
yellow? Might they not be formed from yellow as a base
by the addition of other pigments, or are they entirely inde-
})ehdent of it? Although the two colors are so frequently
associated with each other, the writer is of the opinion thai,
primarily, they were distinct and inde[)endent, but that
yellow also may be, and in fact often is, produced from
brown by oxidation or other chemical action. An evidence
of this is seen in Oiicidium longipe,^, where five of the six
segments of the perianth are chestnut-brown, but the lip is
a rich golden yellow. The lip of orchids is the most
changeal)lc in color, and as a rule has a different hue from
the other petals, and indeed is often variegated with several
colors. Crotons are foliage })Iants remarkable for the
changing colors of their leaves. If the recent leaves are
Hrst green, they are liable later to turn to a bright yellow
along the veins and in spots and patches ; with more age,
the yellow areas become scarlet, and at the same time, and
doubtless by the same pigment or chemical agency, the
lighter green of the lower surface of the older leaves changes
to a chocolate brown. What was green becomes directly
brown. Thus brown is formed without yellow, for no
yellow has been developed there. It is quite evident in
this example that some form of red, nn'ngling with the

12

yellow, produces the scarlet color, and it is found that the
same red with the green produces the chocolate.

A test was made of ordinary pi_o:ments by selecting
Hooker's green No. 2 and crimson-lake, which, mixed in
proper proportions, gave a chocolate identical with the color
of the Croton leaf. We would not presume for a moment
to limit the power of Nature to produce a brown by one
method alone, her resources being unlimited, but in the
present instance it seems to be a satisfactory origin of the
brown, for the color is firs^t met associated with green, an
earlier color, and not with yellow, as may also be seen
from the following :

Epidendrum macrochlhim album, green ground, covered
with red-chocolate, lip white.

Phajus Manni, deep, bronz^^-brown, bordered with green,
lip white.

Anr/ra'cum caudatiim, brownish-green, lip white.

About ten species of Cypripediums, known as Seleni-
pediums, are various shades of green and reddish-brown,
green being generally the ground color. Seven or eight
species of Zygopetalum are green, marl)led with brown or
purplish-brown, lips white or i)urple.

Onddium Lucochilum, green, barred with chocolate.

Brassia lanceana longissima, green, blotched with brown.

Vanda Parishi and several other species are greenish-
yellow spotted with brown.

Catasetum m,acroca')'2)}tm, greenish-yellow, covered with
chocolate-brown.

(Jymhidium canaUci datum, brown with green margins.

Euoninms Eurojxaus, occasionally seen in lawns, has
tiowers perfectly green in color, while E. afrojJ'itrjmreus, the
Burning Bush, much more common, has tiowers of a choco-
late-brown ; the latter color might be formed from the tirst
mentioned by merely the addition of red, and the brown
colored C(d;i/can(hus was doubtless also produced in the
same manner.

13

In GaUlardia tho centriil florets are always green to dark
purple- brown.

JVuphar advena, Yellow Pond Lily : the three outer sepals
often have a reddish-brown band or blotch on the upper
side, appearing even in the bud, through the green color,
before any yellow is developed. In some specimens the
sepals are almost wholly of this chocolate color, which may
have been the original color; at all events it is perfectly
obvious that the brown color is earlier than the yellow.

When green and l)rown are the primary colors, by the
changing of green to yellow, an exceedingly common color
change, we obtain yellow and brown ; then Nature by
some secret art often transmutes the brown into yellow,
one of its component ))arts, by eliminating the others, with
the result of a wholly yellow flower. Thus as regards the
relations of yellow and brown to green, it is found that the
one is as close as the other, that brown in many instances is
the older color, that when brown is a self-color on the flrst
departure, then yellow must rank as second.

In Chrymnthemums the central, tubular florets are yellow
as is commonly the case in Composites; the rays in the
wild flower are usually white, a color which in this instance
evidently was preceded by the yelloW' color of the disc
flowers, from which the rays were developed. In examin-
ing a semi-double Chinese flow^er, one may occasionally see
among the tubular florets the ligulate form in various stages
of development, short or long, flattened out partially or
wholly, but whenever on the disc this altered flower is seen
among the tubular ones, and in whatever stage of develop-
ment, the color is found to be changed to white !

A change of form is often the occasion of a change of
color. These transformed florets, now all white and strain-
shaped, not infrequently become tinted a flesh color, a pink,
or some form of purple or perhaps blue. This sequence of
colors might be farther expanded to embrace the following :
yellow, white, flesh color, pink, rose, carmine, crimson.

14

pink-purple, lilac, l)lue-purple, violet and violet-black.
But this series does not comprise all the well-known hues
of this many colored flower. There are, for instance,
varieties of Chrysanthemums as large and as highly
developed every way, yet having the entire head com-
posed of yellow ligulate petals. Whence then comes this
yellow^ variety ? Simply from the fact that Nature is not
confined to one inflexible law in regard to the order of
development of color, whether in flowers of simple or of
complex construction.

If one examines the central florets of a semi-double yellow
flower, it will be seen that upon the change of form to ligu-
late, no change whatever is effected in its color, and thus
the entire flower in its transformed character from tubular
to ligulate remains yellow, not losing that color as in the
former series, but on the contrary is very apt to add to it
some form of red or even of purple, resulting in many of
the most beautiful secondary and tertiary hues known to
flowers. This scries would embrace : yellow, straw, orange,
scarlet, salmon, purple-brown, maroon and black, and inter-
mediate hues with an infinite variety of tints and shades
difficult to describe. That this last series of color is not
confined to high classed and cultivated flowxrs can be shown
by going back to simple Ranunculacefjn, than which none of
the polypetalous division are more primitive ; and among
the genera, Clematis may be regarded as jierhaps the sim-
plest. Here are but a fcAV yello\y flow-ers, none wholly
yellow in North America, yet Clematis reticulata, Walt, is
said to be pale yellow inside, reddish outside. C. coccinea
is scarlet. Aquilegia Canadensis, sometimes yellow, is
generally scarlet, except at the tip and within the petals,
where it is yellow; A. fiavescens has "sepals sometimes
scarlet tinted outside."

Trollius has two species, deep orange-colored.

Potentilla Wvi- Rollisson is orange-scarlet, a simple
flower of liosacea'. liammadus, or buttercup itself with all

15

its claimed simplicity, adds rod in 7?. Asiafinis iVOlamht;
in cultivation it is a magniticcnt double flower of a bright
scarlet color !

It will be observed that this change of color is in the
same direction in the simple Clematis, TroUiusi and Raimii-
cidns, etc., as in the more specialized Columbine and in the
highly developed Chrysanthenmm, and that the colors do
not pass through white to pink, etc., which fact will be
especially referred to farther on.

It may also l)c added that the simpler Clematis has more
purples and blues than the higher developed Aquilegia, or
at least as many, the latter having ten yellow or scarlet spe-
cies to only two purple in North American species. The
list of flowers belonging to this series, many exclusively so,
is immense. Among the number are the following, in
several of w^hich the red pigment may actually be seen flush-
ing from day to day the yellow surface of the j^etals :
CupJiea, Gorezema, Manetlia, Libonia, Gesneria zehrma,
Canna, Lantana, Tropcmlum, Tulip, Poppy, Tritoma, Es-
choUzia, French Marigold, Lonicera sempervirens, LachenaJia
Nehonii, Rosa Inlea, var. Copper Austrian Brier ift yellow
externally, but inside a brilliant orange-red. Taraxacurii
(Dandelion) and Leontodou, have the outer row of ray
petals, commonly a reddish-purple on the lower surface.
In upwards of one hundred species of Aster, the central
yellow florets change directly to a reddish-purple.

Polyr/dla sanguinea and P. cruciata, have an exceedingly
interesting color change not noticed by any of the Floras :
The conspicuous color in the heads of these flowers, usually
a rose-purple, belongs to the sepals, Imt the small true
flowers are first yellow, then orange, and Anally a rose-
l^urple, very much like the sepals in color. All these colors
may be clearly seen in the same head in a bright, clear day !

CaMilleia coccinea, flowers, or rather the bracts, yellow
and scarlet. ''Garden and Forest" states that flowers of the
yellow American Hamamelis have been found in New York

16

state with bright red petiils. From the foregoing it will be
seen that yellow to red is a very common and natural suc-
cession of color, especially if the red be a vermilion or
scarlet. Kni])hofia or Tritoma, the Flame Flower, retro-
grades from scarlet to yellowy the young flowers at the top
being scarlet, and the older at the lower end of the spike
being yellow.

In view of the above well-established facts regarding the
order of color change, it is truly astounding that Grant
Allen should make a statement so glaringly erroneous as
the following, which ignores nearly one-half our choicest
and best known blossoms :

"As flowers advance in type the}' pass from yellow, which
is the lowest color, through white, pink, red, and lilac to
purple and blue, which are the highest. And wdien through
any special cause they begin to retrogress, they pass back-
ward through the same stages in inverse order" !

Grant Allen recognizes two compound colors, lilac and
purple, in his scale, but utterly ignores others, such as
orange, scarlet, brown, brown-purple, green and other of
the second series above mentioned, and his sequence is one
only of two wdiich are equally imjiortant, not mentioning
others of less occurrence.

CHAPTER II.

Certain yellows evidently are the product of a i)receding
chlorophyl green. Chemists inform us that the coloring of
chlorophyl is composed mainly of yellow and blue pig-
ments, which can readily l)e separated.

This process takes place naturally in leaf and petal, the
l)lue portion ))eing absorbed by the plant, or in some other
manner eliminated, while the yellow element remains. This
may be illustrated by the Chinese primrose, which, of what-
ever hue, has a central green eye. This green spot often
remains unchanged, but usually it becomes moditied, in
color varying from yellowish green to pure yellow, the
yellow color being confined strictly to the space previously
occupied l)y the green. It is obvious therefore that this
particular yellow was occasioned by the decomposition of
the chlorophyl. The same is true in most cases of both
green petals and green leaves, viz., that the yellow depends
for its existence upon a previous green color. If one ob-
serves the ripening leaf, even before there is any frost, say
as early as the middle of August, he will tind that some of
those of Rhus copalinu, for instance, have become already of
a yellow color, other leaves on the same bush will be scarlet, *
while others still may be partly yellow and partly scarlet.
A similar color change is seen in maples, Crofon, Coleus,
AcJiijvantJtes and in green petaled liowers. Which color
ought one to i)ron()unce the most primitive? According to
Prof. Vines, "the green and the yellow coloring matters of
|)Iants are deposited in corpuscles of a protoplasmic nature,
but the others, especially the red and l)lue, are dissolved in
the cell sap. It is to the presence of coloring matters in
solution in the cell sa[) that the colors of petals are princi-
pally due." The red and blue colors then are quite inde-
pendent of the yellow. They may develop before the

18

chlor()i)hyl has dissolved into its components, or tiiey may
appear while the decomposition is taking place or after-
wards, or the yellow principle may vanish with the chloro-
phyl and so never develop. The red coloring matter suf-
fusing the green or the yellow petal or leaf produces a
scarlet or a darker shade, precisely as the mixing of artists'
pigments would do. Therefore it is apparent that when a
chlorophyl green is the foundation, the first color to follow
is not restricted to yellow, but may be red, purple, or blue.
There are other yellows, however, or at least of a differ-
ent origin, which are developed from a pure white founda-
tion. These are seen in Lonicera longijiora, garden bush,
pole and Lima beans, many orchids, and in a less degree in
tuberose, jasmine, etc., which become yellowish by age,
and in Fungi.

We know furthermore, by the examples heretofore given
of Cypripedium and Angrtecum, that some greens do not
become yellow at all, l)ut they gradually and sometimes very
slowly indeed, as in LapiKieriu alha^ fade evenly away to a
pure white ! *

Therefore not only a yellow and a green color, but for
that matter any color, to use a popular ex[)ression, may fade
out to a pure white, as we may have occasion to show.

The writer has not yet Ijeen so fortunate as to have seen
the original statement of De Candolle regarding color. If
his oft quoted "xanthic and cyanic series" referred to spe-
cies of flowers, each of these divisions would contain the
most heterogenous assembly of flowers imaginable, and such
an arrangement could hardly be conceivable of a botanist of
his intelligence ; he doubtless referred merely to the evolu-
tion of color as a separate and indei)endent featui'e of
flowers, such a division l)eing not inappropriate, for, intleed,
with com})aratively few exceptions, all hues might be com-
prised in these two series to which we have already alludetl ;

* See LapaEjeria alba aud otliers iu the list, Chapter IX.

19

the xanthic coinprisinu' green, yellow, orange, scarlet, etc.,
etc., all containing some yellow, and therefore making the
designation quite a})propriate, while the cyanic or blue
series, beginning as in the xanthic with green and yellow,
would diverge at this i)oint and continue through white,
red, crimson, pur})le and violet to blue ; or white might
immediatel}^ follow green, in which case yellow as a rule
would be excluded from the series.

Numerous browns, brown-purples and similar shades
wt)uld not be in either direct line, but might be considered
as early digressions from the regular gamuts.

Although the progression of each series is given from the
lighter to the darker tints, it should be remembered that the
reversed order would be just as natural and just as primary,
as Nature has no hard and fast rule regarding color'. Each
series also is stated in a definite sequence, not because it is
an invariable rule, but because ])roken links of the chain
conq)r4sing two, or occasionally three colors, may be found,
which united will make a complete series.

To illustrate, there are many si)ecies in Borraginacea\
having flowers which expand of a pink or light crimson
color, but (juickly and regularly change to })urple or blue :
and white flowers are very a})t to become pink, while the
white color itself fre(]ucntly can. be traced to a previous
yellow ; so that the order is yellow, white, red, pur[)le.
This succession of colors may occasionally l)e noticed in a
single flower, as in the (.'hinese primrose ; it has a green
eye, changing more or less distinctly to yellow, the other
part of the flower has become white quite likely, but not
necessarily, from a previous yellow like the eye, the latter
being the honey-guide relfc which will be particularly dis-
cussed in the second chapter. This same white flower often
turns crimson and in fading becomes a blue-purple. But
while this order ap[)ears to be very connnon, as a matter of
fact every color may spring directly from white. Scores of
species are known only as white or blue, white or red.

20

white or yellow, each of the latter colors proceeding directly
from white or retrograding to white ; or these three pri-
mary colors may, as we have shown, originate immediately
from the chlorophyl green, or actually precede it !

CHAPTER III,

The lianumnilaced' are i)laced first in order in our Floras,
Ave imagine for the reason that the fiovvcrs of this family
are very simple in construction, especially so if we except
some peculiar shapes of petals in Aquilegia, Aconitatn and
Delphinium, but even in these as well as in all the others,
all the parts are separate and distinct. Hermann Miiller
calls a flower simple which is without any consolidation of
either the essential organs or floral envelopes.

Grant Allen, in elucidating his color theory, selects from
among this family the common yellow buttercup) (Ranun-
culus) as a typical example of a simple or primitive flower
and maintains that it is prevailingly yellow "because it is
an early and simple type of flower."

But before accepting his conclusions one should be fully
assured as to the correctness of his premises ; we therefore
would like to be informed in the first place, if it really is
"the simplest and least differentiated member of the group,"
as he puts it, and if there is no member equally as simple
and yet having a different color.

Can one believe that there are no species of that family
even more primitive in character? Take for instance Cltm-
rtds Virginiana, the Virgin's-Bower ; this flower has not even
developed any petals, it is so primitive, and the number of its
white sepals is but four, instead of the more common number
five. A double Clematis srowinu" in the writer's grounds

21

throws some light on the color of tliis flower and on the
cause of its having hut four sei)als. It was renuirkahie in
having its leaves near the flower which was terminal, white
like the flower. The lower leaves were oi)posite, those
higher u}) in whorls of four, or two pair hrought together,
while near the flower, a whorl of twelve leaves, that is six
pairs, was completely merged into sepals. The four outer
had long petioles with blades, mostly green, hut containing
some white, a second inner row of four had more white on
the blades with shorter })etioles, while within these the
blades and claws, wholly white, were of the same length as
the sepals or real flower. Some of the stamens had the
anthers unrolled into small petals, su})ported by long fila-
ments ; thus there was a conn)lete gradation from foliar
leaf to petaloid stamen. It was obvious that the white
color from leaf to stamen resulted from the same cause,
viz., the lack of chlorophyl. If any color could in reality
l)e more ])rimary than another, the white or greenish-white
of (/. Viryiniitna, being at least nearer to the fundamental
tissue color, is entitled to that honor

From the above example of an abnormal Clematis we
learn that the four sepals are merely a whorl of four leaves
sim|)ly reduced in size, ;ind not being any longer vegetative
in character have dispensetl witli their chlorophyl, and now
white, may assume sonu' other color as purple or l)lue, if
such color should be for tiieir advantage. As a matter of
fact, the different s[)ecies are almost wholly white or purple,
there being very few ind(!ed of a yellow color.

There are perhaps a dozen species in cultivation of a
white color. The anthers, at least in some of them, are not
even yellow, but as white as the filaments and sepals, and
moreover, so slightly differentiated from the hlamenl as
scarcely to be detected l)y the unaided eye. The })ollen
also is white !

It is not necessary in these flowers to consider the <|ues-
tion whether original petals are derived from stamens or

22

not, for we know that the colored sepals of Clematis never
were stamens ! If we compare Clematis with Ranunculus
we find that the former in the absence of true petals is cer-
tainly more primitive, or at least more simple, than the
Buttercup, which is a complete flower, being furnished with
both whorls of floral envelopes, and the thick, glossy petals
still farther advanced by the presence of scale-like necta-
ries at their base.

A little more developed than Clematis, are Anemone p((tens
(Pulsatilla), and Atragene, both of which make a feeble but
vain attempt to produce petals by their "gland-like, abortive
stamens answering to petals," and by "enlarged filaments
more or less petaloid."

Still higher perhaps in the scale of development is (Jiuii-
oifuga, which actually has a few diminutive petals, which
with the numerous stamens are white !

We might allude to the frail, purple-flowered Anemone
nemorosa as another blossom much simpler than Buttercup,
and also refer to the fact that of this genus A. Robinson-
iana, A. apennina, A. blanda, are of a sky-blue color!
But we are not obliged to go out of the Ranunculus genus
itself to find real Buttercups having white, red, or scarlet
flowers as well as those of a yellow color.

R. Andersonii, Great Salt Lake vicinity, is i)ink I

R. Asiaticus has yellow petals with reddish tips, stamens
with the anthers and the [)istils brown !

R. Asiaticus var. d'Olanda, a cultivated variety, is a
magnificent double scarlet flower as large as a double
Dahlia ! Some of these flowers show a little yellow
at l)ase of the petals, thus proving that the color was
formed directly from yellow by the addition of red.

R. Anemonoides is white tinted with i)ink !

R. Li/alli is pure white, stamens yellow 1

There are varieties of R. Asiaticus which are purple !
yellow, orange, and variegated white, yellow or red; thus
all colors except l)lue are already found among Buttercups

23

and it would not be so very astonishino- if a l)luo ()iu> should
.some day be discovered.

A partial list of Hanwiciikfcerc, showing their colors :
Ilelleborus colchiciis, bright })uri)le, stamens yellow.
JVigella Hispanica, blue, stamens l)lood-red.
CimirJfuga racemosa, every i)art of the flower is white,

including filaments and anthers.
Clematis caermla, violet, stamens deep purple.

Do. Vi'rrjmicniu, white, stamens white.
Hepalica triloba, lilue, filaments and anthers white.

Do. angidosa, l)lue, filaments, anthers and [)ollen white.
Ranunculus Lyalli, pure white, .stamens yellow.
Do. Austijiea, pure white, stamens yellow.
Do. Asiafinis, various colors, stamens brown.
Do. ljulbosi(s, etc., every part yellow.
( Atragene) , Clematis alpina, bluish-violet, stamens yellow.
ThaJictrxim Cormiti, white, filament and })istils white or

greenish.
Impyrum, Acta'U alba, Clematis of white flowered species,

have filaments and anthers white.
Anemone Virginiana, every part white.
Anemone nernorosa, white or purple, filaments and anthers
snow-white.
Do. Italian, scarlet or i)ur[)le, anthers and stigmas
dark purple, almost black.
Tranvetteria, all jiarts white.

Nigella, Xanthorrliiza, Aconifum, blue or purple, and
Anemone Virginiana, white, have green or greenish
stamens !
Delphinvm Ajacis, })urple, filaments and style whitish,
anthers black-purple.
Do. tri.'ife, dark, dusky brown.
Anemonella tltalidroides, white, filaments white, anthers

yellow.
Aquilegia Canadensis, scarlet and yellow, filaments pale
green.

24

Acorn'fvm CaJifornicuni, bliio-purplo, filaments and anthers
green .

Hydrastis Canadensis, greenish-white, tihinients and an-
thers white.

llellehorus vtridis, purple, tubular petals very small, per-
fectly green.

Oaltha, yellow, white or pink.

Clematis, white, scarlet, |)urple, greenish, and a few
yellow.

Helhhorus, purple, greenish-purple and white.

Aconitum, purple, blue, white, yellow.

Anemone, white, greenish-white, purple, pink, sulphur
color, blue, yellow.

Adonis, yellow or red.

Ranunculus, yellow, white, pink, scarlet, or partly purple !

Delphinium, scarlet se{)als and yellow petals, purple,
l)lue, white.

Among a dozen species of American Aquilegia, three-
fourths of the number are yello^v, notwithstanding their
high development.

Aconitum has many blue flowered species, but it also has
at least thirteen species or varieties of a yellow or cream
color !

We have shown that the prevailing color of numerous
species unquestionably simpler than the buttercups, are
white or purple, with several reds and many blues. The
remarkably developed Aquilegia is mainly yellowy and
while Delphinium and Aconitum have in the American
species the most blues, there is really very little difference
in colors between the species of the highest and lowest of
the Crowfoot family, or say between Aconitum and Clematis.

There are to be found among the Iianunculacew an un-
usually large number of sjiecies having only colored sepals,
except that at the same time there is apparently an attempt
to produce real petals, with the result only of modified sta-
mens of the outer row, which do not arow at once as in the

25

case of doul>lc flowers into perfected petals, hut remain half
transformed and diminutive.

They represent at least a transition state, and being the
nearest approach to true i)etals, ought to l)e of a golden
yellow if it be true that all petals were originally of that
color, but in reality they are generally white, also the sta-
mens from which they develop are white I

We have discussed the Bammcukicea' at some length, for
the reason that great stress was laid on the yellow buttercup
by the author of "The Progressive Color Theory," as show-
ing an example of a simple flower having "the golden yellow"
color said to be peculiar to "almost all of the earliest and
simi)lest tyi)es of cxish'n;/ flowei-s," which are seldom white
and never blue. ' I

CHAPTER IV.

It might be interesting to notice a few more statements
of Grant Allen, who endeavors to show why the buttercup
is 3'ellow, the stitchwort white, etc., viz. :

1. "All flowers it would seem were in their earliest form
yellow ; then some of them l)ecome white ; after that a few
of them grew to l)e red or purple : and finally a com}>ara-
tively small number acf(uired shades of lihic, mauve, violet
or blue, and when through any special cause they i)egin to
retrograde, the}' pass backward through the same stages in
inverse order." "Almost all the members of the most ad-
vanced families are purple or blue."

2. "The Violacece are a whole family of bilateral flowers
highly adapted to fertilization by insects ; and as a rule they
are deep blue in color. This is the case with four of our
British species."

26

3. "The highest mode of adaptation to insect visits is
found in larkspur, Delphinium ajacis, which is hhie, white
or red; and still more develoi)ed in monkshood, Aconitnrn
iiapellus, one of the deepest blue tiowers we possess."

4. "Aqniler/ia vulgaris is bhie or dull purple, hut readily
reverts to white or red."

5. "The CoroUifora' betoken in their shape high modi-
fication ; yellow is a comparatively rare color, while purple
or blue become almost the rule."

6. "The Borrar/inacea' are another very advanced family
of Corolliflonv, and they arc blue almost without exception."

7. "Most early and simple flowers are yellow because
the stamens are generally yellow, and when they developed
into petals they naturally retained at first their original
coloring."

8. "In ox-eye daisy and May- weed the rays have become
white, and this I think establishes the fact that white is a
higher development of color than yellow."

9. "It is not remarkal)le that the pinks should never be
yellow, as the five principal carpels have completely coa-
lesced into a five celled ovary."

10. Progressive coloration is said to follow the modifi-
cation of flowers from the simpler forms to those highly
specialized. Among the former are buttercups, potentillas,
the AlsinecM, and Alismaceoe, as Alisma and Sagittaria ; and
among the latter, violets, peas, composites and orchids,
harebells, heaths and labiates.

We answer in the same order :

No. 1. Has already been for the most part fully an-
swered. In Ranunculacecv there is not the slightest evi-
dence that the numerous purples and blues have ever been
yellow, and as may be seen from a foregoing list, the sta-
mens are by no means prevailingly yellow, and from a list
a little farther on it will lie proved that purples and blues
are no more common in orchids, Legvminosa\ Compofiitu',
Scrop)hularia(-ea', etc., than in Banunculacea', etc.

27

No. 2. Every one knows that in our Northern States
white and yellow violets are eomnion, there heini*- four or
five white tiowered spceies and three yellow tlowered s[)e-
eies, with several other yellow kinds in California.

No. 3. Delphinium has two or more species with yellow
petals, and Aconitum thirteen species and varieties yellow
or cream color !

No. 4. (^ur New Enijland Aquilegia is scarlet tipped
with yellow ; it is not a reversion, l)ut a pro<2:ression fiom
yellow to scarlet. California has two or more species
wholly yellow !

No. 5. We have only to point to the immense number
of yellow disc florets of this form of flower in Composites,
to the squash, cucumber, (lerardia, 3/imiilns, Verhasoum,
Calceolaria, Oenothera, Lonlcera, AUamanda, Bf(/)io)iia,
etc., etc., all yellow^ 1

No. (5. We count in Lowden, whose list is very incom-
plete, thirty, white species and twenty yellow flowered
species !

No. 7. The stamens in liannncuJacea' are very largely
white or ])urple.

No. 8. In Cineraria, the cultivated species are mostly
crimson, purple or blue and never yellow, and the centre
florets are of the same color ! These rays in various varie-
ties are in all stages to })ure white. Serioocarpiis eonyzoidex,
central florets pink-purple, rays white. Nearly every color
under favorable circumstances becomes white

No. 9. The great mjijority of species have their carpels
more or less united ; we will only mention a few : Nuphar
ovary, eight to twenty-four cells, yellow : (Jenofhera pod,
four cells, yellow ; Hypericum pod one celled, styles three
yellow, etc., etc.

No. 10. We have shown that buttercups have various
colors, and we can add that Potentilla has besid(^ the com-
mon yellow also rose, white, scarlet and dark crimson hues,
and in P. palustris a purple color !

28

In regard to Alismaceo' we quote as follows :

"The common arrowhead is one of the earliest and sim-
l)lest threefold flowers of separate sexes, with a white
corolla of somewhat papery petals. The water })lantain,
another form of early threefold blossom, ])oth sexes com-
hincd, has delicate pinky-white })etals and a numl)er of
small one-seeded carpels exactly as in the buttercup, which
occupies in part the corresponding place among the tive-fold
flowers. Save that the petals are now pinky-white, while
those of the original ancestors were almost certain] // yellow,
we might almost say that the marsh-weed in question was
really the earliest petal-bearing plant of which we are in
search . "

AVhen Grant Allen writes on the histoiy of certain
plants and does not treat specially of their color features,
his observations are at variance with his favorite theory, as
seen in the colors of Alisma and Sagittaria. liy his theory,
as the Alisma is one of the earliest and simplest flowers, it
ought to be yellow, l)ut it is really white or even pink,
without the slightest trace of yellow. This discrepancy is
noticed by its author and is remedied by hypothesis : by
guessing that sometime in the long past, it might l)e 80,000
years or so, the flower "almost certainly was yellow" I

It is indeed very unfortunate that not a single ex;miple
of any existing flower could be found in all the great endo-
geneous class of [)lants to illustrate the theory !

If all petals and stamens were first yellow, why not the
pistils and the base of the pistils, which later becomes the
frnit, as this is but a part of the flower which continues to
develop after certain other parts wither and decay.

i)oul)tless this idea would strike every one as absurd, but
it is quite as reasonable as that every flower was originally
yellow.

The develoi)ment of color in fruits from the green state
to maturity is a very interesting study, and the methods arc
as diverse as one can conceive.

CHAPTER V.

Changes which take place in the individual flower, prov-
ing that THE SEQUENCE OF COLOR IS NOT Uniform.

Viola ti'k-oloi\ yellow to purple or blue, and purple to
yellow.

^senilis and Catalpa, honey guides yellow, change to
rose-purple.

Pehirf/oniiit7i, scarlet varieties are a))t to be rose-color in
bud.

Aster nmbeUaliis, white ; central florets yellow to pale
green.

Do. corymbosus, white ; central florets yellow to rose-
purple.

Of the large family of Asters, one hundred and fifty spe-
cies, more or less, the central florets are usually yellow at
flrst, but change regularly by age to some form of purple.

StrophoMyles anr/ulosa, the green bud expands a clear
pink color, changing, except the diminutive tip of
the keel, entirely during the day to a creamy yellow
or light l)uff in the evening.

Pliaseolus vulgaris, common garden varieties either wliite
or purple and the Lima bean, change color to yellow,
especially the wing petals.

Ilelianthus cncmuifolius, yellow : central florets with yel-
low" tube, the lobes only, dark, red maroon, chlorotic
chaff among the florets tii)})ed with crimson.

Cassia nil-titans, yellow ; stamens yellow at base, upper
half is crimson, changing to purple and the anthers
purple I

Oleoma, a variety with slender white petals on long pur-
})lish claws, filaments purple, the petals have a tend-
ency to become a lemon- vellow,

30

Globe Amaranth, the head is composed mainly of the
prominent dariv crimson bracts ; the minute flower
is first white, but changes to the same crimson.

Zinnia has large, coarse textured outer petals ; some
varieties of these at first lemon yellow, become
orange, then scarlet, and with age (crimson !

The central florets do not change their yellow color. Cer-
tain other varieties begin with white which is then followed
by quite a different series of color, similar to those shown
in Chrysanthemums.

Prenanihas alha, white, has a light green involucre or
sepals which change to purple.

Chrysantheimim tri-coloris variable, the rays being yellow^
at base and red at the tip, or yellow at base, then
red above and the tip white, while the disc florets
are })urple-brown.

Gheirantkes Cheiri, certain varieties at first lemon-yellow,
change directly to pink.

Phaseolus vulgaris var. mnltifiorus. The inner petals of
a very young bud of the Scarlet Runner bean are
pink and white ; when half grown and when ex-
panded they are orange-scarlet.

Euphorbia pohjgonifolia and E. iuacidafa, white, on ex-
panding change to pink.

Sap)onaria officinalis, frequently ex}>ands white, after-
wards changes to pink and fades finally to purple.

Polygonum aviculare, the green se})als have a white bor-
der changing to pink.

Arctium Lappa^ burdock, not infrequently t)i)ens white
and changes to rose-pur[)le.

Gaillardia, central florets green, tipped with purple-
brown approaching crimson on the outer row. The
rays also tubular, though much elongated and with
modified lobes, are greatly changed in color from the
central florets, as the tubes are dull scarlet and the
lobes bright yellow ! This is a good example of

31

different manner of color development and change in

the same compound flower. The yellow here is

secondary to the red !
CakUe Americana, commonly at first white and its fila-
ments green, both change to purple, and the green

sepals with age become yellow.
Rhexia Virginica, rose-purple, the yellow filaments change

to scarlet after expanding, also the lower half of the

yellow anther.
Sisi/rinchiuiii auguMifolium, blue, the u[)per portion of

the yellow filament changes to blue, while the lower

part remains unchanged !
Poly gala iwhjgama and P. cruciata have their true flowers

pale yellow, changing through orange to rose-purple !
Cohea scandens, pale green the flrst day, the inside of the

lobes become pur[)lish on the second day, and by the

third day the entire bell-shaped corolla has become a

be-iutiful purple !
Erigeron Canadensis, the outer rows of minute white

flowers turn purple.
Lindelojia, Symphyluni, Anc/tusa, Om.2)holodea,Mertensia,

Pidmonaria, etc., change from pink or red to blue.
Myosotis laxa, in bud white, becomes first pink, then

light blue. The different species all have a yellow^

eye.
Gera.rdia uiaritima, petals purple expanded, are of a

pale yellow in a half developed bud. Anthers and

pollen snow-wdiite.
Lycopiis Virginicus, white ; in the bud or just before

expanding is a pale yellow ; at least those growing

in dry soil and sunny places ; also Alyssum mariii-

mnni, white, under certain circumstances has been

observed to be pale yellow in bud.
Aster vimineus, white, and several other white species

change to pink, lilac or rose-purple.
Primula Sinensis, Chinese primrose, has varieties first

32

white, chaiiffino; through pink to crimson or y)urplo and
fading to blue-i)urple.

Lonicera longiflora, i)ure white changes to pure yellow I
and several other species of different colors become
decidedly yellowish with age.

Loniceras have a strong tendency to yellow as seen by
the following descriptions : cream colored, yellowish-white,
greenish-yellow, greenish-yellow tinged Avith dull purple,
yellow tinged with red, scarlet outside and yellow within,
scarlet, whitish with a purple tube fading yellowish, at first
snow-white but finally changing to golden yellow I

Composite flowers, having different shaped flowers in the
same head have different color changes : Bellis, the daisy,
Chrysanthemvm, etc., change the central, tubular, yellow
florets to white, ligulate ray petals, which often become pink
or in some genera purple or blue ; other genera change the
tubular to ligulate without change of color as in Rudbeckia ,
Coreopsis, Bnhhvinia and Helenivm, but do change the
central florets to brown or brownish-purple ; in others the
yellow rays change to some of the following colors : l)rown,
brownish- red, maroon, copper colored or dark red-purple
as in Coreopsis coronata, G. Ilarvei/ana, Heleniwn elegaits
and 11. midijiorufn and in several species of GaiUardicr,
Df/sodia, Psathyrotes, Troximon, Lactuca, JVabahts, etc. ;
in Erigeron bellidifolius, the central florets turn a rust}'
color ; in Sericocarjms con yzo ides the central florets are
never yellow, but pink-purple usually changing to white,
while Cineraria, at least in ordinary varieties, has both
rays and centre of same color, viz. : crimson, lilac, purple
or blue, and never yellow I Asters have rays of different
colors from their centres, which last commonly change
from yellow to a vinous purple.

Even when the two kinds of flowers in the same head
have similar colors, the variegation in some species is
exactly reversed, as in Gaillardia and Coreopsis, the purple-
brown color being at the tip of the yellow corollas of the

33

centre and at the base of the rays ! Hieracmm aurantiacum
<>-oes from yellow to orange. P(frethrum uh'gmosum, ray.s
white, has central florets of a green color, the minute lobes
only are tipi)ed with black I

Florists' varieties of Chinese Chrysanthemums, Dahlias,
etc., have innumerable hues.

CHAPTER YT.

Where yellow flowers niaj' be found :

Helleborus, several species absolutely green, and NigeUa

Nigellastrum, flowers brown and green I
Aqailegia has several handsome bright yellow species,

and notwithstanding that this flower is highly spe-
cialized. A. viridifora is green flowered I
iJelphlnium has one section consisting of species with

calyx and petals of different colors, the petals being

wholly or partly yellow.
Acoiiitum, in section of flbroiis roots, thirteen species or

varieties are yellow or cream color, only six ])urple,

but in section of tuberous roots are mostly blue or

violet.
Anemone and Hepatka, very simple flowers, mostly blue,

purple or white.
Clematis, chiefly l)lue, jxirple or white with very few

yellows generally of a dull color.
Ranuncvhis, mostly yellow, has many whites, also pink

and purple.
Paeonia, though of the llanunculaceua, appears to have

no 3'ellow flowered species !
Caltha, conmionly yellow, has species rose tinted or

bluish.
All of the al)ove belong to Ranunculace;e.

34

Malvacece: Abutilon, eight species cultivated are yellow
or orange.
Malva, twelve yellows and many purples.
Hibiscus, twenty j^ellows and twelve purples.
Sida, tifty-eiglit yellows and one })uiple

AmaryUiacem : Alsfroineria, Bomarea, Zeplri/mnlhei^^
Narcissus: this last has tifty-four species either
white or yellow, some with highly developed cups or
trumpets, also the others very handsome, preponder-
ating colors yellow, orange or scarlet. Hypoxis
same order, all yellow.

LiJiacece: Blandfordia, Heine rocallis , large, beautiful,
chiefly yellow, larger and handsomer than Funkia,
white or purple, also Lilium, Erythronium, Uvula-
ria, TuUpa, Kniphojia, etc.

Bigiioniacem, Bignonia fiowers tubular, ten cultivated
yellow species, also Tecoma, etc.

Scrojjhulan'acea' : Veronica has eighty species wholly
purple ; this genus has its color ascribed to it on
account of high development ; but high development
is wholly valueless with regard to color, with Ver-
bascum, Celsia, Gerardia, Bedicularis, Liuaria,
Verbascum, Mimidus, Gratiola, and many other gen-
era of the same family, largely yellow and superior
every way. Calceolaria with yellow or xanthic hues
is more highly modified.

Ericacea' : some of the largest and handsomest, as in
Rhodendron, are yellow.

Hi/pericace(V : Hypericum, sixty-six yellows and one red !

Onagracea' : Oenothera, a large family, is almost wholly
yellow,

Primulacea' : Primula, a dozen species yellow colored,
tubular, salver shaped, etc.
Lysimachia and Steironema, all yellow.
The smallest flowered species have the darkest
hues !

an

Crucifera' : the flowers are about c<|iuilly divided into
white, yeUow or purple colored.

(^enuilaced' : Pelargoniuu), thirteen yellow species, hii-
l)atiens and Oxalis, Tropteokun, etc

liitbkicea',: Palicourea, a genus of one hundred species
with elongated tubes, mostly yellow or white, never
blue I Rondeletia, many yellow and white ; yet
Houdonia of this family has thirteen species and
varieties, small flowered, all blue or purple. Boh-
vardia has some yellows.

Liihiatw has many yellows in Sideritis, (kdeopsis, G(de-
obdohn, Sfach//s, PJdomis, Leonotis, Teucriuiii,
Ilyssopiia, /Salvia aurea, S. glntinosa, iS. indica, )S.
officinalis, etc. ; on the other hand Mentha and Thy-
mus l)ear small, puri)le flowers and not a single
yellow one ! On the whole, purple and blue far
exceed all others.

Iridem: Gladiolus, Iris, Ixia, etc.

(Japrifoliacem : Lonicera, a dozen or more yellow flowered
species.

Canjophyllacece, : yellow flowers are scarce, but Silene has
three s})ecies, Saponaria one, and Dianthus one, and
doubtless there are others.

Cucitrbitacea' : largely monopetalous, and even bell shaped
are quite generally yellow as in pumpkin, cucumber,
watermelon, etc.

Borraginacea' : flowers largely pur})le or blue, or chani>-
ing from pink to blue, yet Amsinckia, Onosi/ia, Lilh-
osperinium section Batsdiia, all yellow or orange !

Rosacem: Rose, the Queen of Flowers, has yellow species
and other genera yellow.

Compo.sita' : native composite flowers have rifty [)er cent,
more yellow flowered than purple flowered species !
This family probably has more yellow flowered spe-
cies than all others of the family put together.

36

Uinhellifera> : have for the most part white tlowers, next
those of a yellow or green color.

Ler/uininosce : the tlowers are recognized at once as highly
modified and specialized, with buttertly shaped cor-
ollas mainly, yet upon an enumeration of the species
in London's Encyclopedia there were found to be
548 yellow s})ecies to 372 of all other hues put to-
gether !

Lentibulariaceoe : Utricularia, largely yellow.

Violariaceoi : violets, about as many yellows as white or
violet, there being two yellow flowered s[)ecies in
Massachusetts.

Liliacem: Aloe, all colors, yet forty-three out of ninety-
nine species are green !

Yerbenace(K : Verbena, a large family almost wholly red,
purple or blue, but the flowers are small or only
mediocre ; Amasonia, however, corolla with elon-
gated tube, sub-bilabiate, has four species yellow or
sulphur-colored I

Orcliidacece: There are very few native orchids of a
yellow color, but many of a green or greenish color,
a few rose or brown, but no blues ! while exotics are
pretty evenly divided among purple, red, yellow,
white or green, with many of a brown color and
very few blue ! The prominence of green flowered
species of both small and large size is quite notice-
able, and yet this family is in the highest rank of
specialization and modification.

Alistnacece: has Butonius and Alisnia lose colored, Sagit-
taria white.

Naiadacece: ranking lowest of all, Apo)ioyeto)i^ no petals,
but large bracts white or pink, anthers })ur})le-
brown.

We are ()l)liged to omit many other Orders for want of
s[)ace, but if one could see a coniplete list of yellow flow-
ered species he would incontinently abandon the idea, if he

37

ever had .such, that yellow flowers are confined to simple or
primitive types. We think, however, that the general list
of Orders above given will be ample evidence as to where
these species are to be found.

White is designated as a low or i)rimary color, yet as in
the case of yellow, it is found all through the tioral world :
Viburnum, Cornia^, Androuieda, Saxifraga, Arenaria,
Cratcegits, Capsicwn, Clevodendron, Umhelliferm, the mag-
nificent Easter Lily, trumpet-shaped, nine inches in length,
Pancrathun, Convalktria, Nijutpliam, Yucca, Gordonia,
Slnarlia, Camellia, Exocliorda, etc., magnificent Orchids
and Chi-ysanthemums, etc., etc. I

We think that enough has been said to make it perfectly
clear that the statement: "that almost all the members of
the most advanced families are i)urple or blue," is very
inexact.

COLOR STUDIES.

CHAPTER yil.

Myosotis, Forget-me-not, presents a succession of four
colors. The stamens and eye arc yellow ; the five minute
rays extending from the eye and completing the star-like
centre are white ; the corolla, at first a clear pink, changes
gradually to deep blue. The order is yellow, white, pink,
blue ; but if one examines a bed of Myosotis critically he is
likely to notice exceptions to this se(iuence.

The yellow anther never becomes white, but changes di-
rectly to brown ! The nectary changes to white only ; not
unconnnonly an entire plant bears pink flowers, changing if
at all to white ; and such flowers as do become blue are
apt to change again, but to white also.

38

Thus all three colors under certain conditions become
white. The common white variety is derived from the
blue, and with a lens, frequently remnants of the blue may
still be detected. It will be observed that these colors do
not retrograde in the same order as they advance.

While many exau)ples of a sequence similar to above are
to be met with, there are certain yellows that change di-
rectl}^ to blue, no white or red intervening, though usually
some white, as in Pansy. The yellow anthei's and pollen
of crimson Tulips with a yellow centre become blue ; and
these yellow centres themselves change through green to a
dingy blue, which in the variety T. Gesneriana is a perma-
nent pure blue which sometimes appears as a blue star on a
round yellow eye I

On certain yellow Pansies, not the lightest yellow, but
those more of a Cadmium yellow, is frequently to be seen a
dingy or olive green blotch which finally disappears, leaving
the flower of a clear color. If one will examine the half or
even two-thirds developed bud of the same plant, he will
find that its color is then partly or wholly violet, and that
the commingling of this with the developing yellow^ occa-
sions the greenish blotch. Sometimes a flower may be
found entirely and evenly colored a greenish yellow from
the same cause. Furthermore, if one picks open the bud
of a purple-brown or reddish-brown and yellow variety,
which is not a rare combination, he will find it also to be
purple. These brown varieties fade finally to yellow. Any-
one therefore can satisfy himself of the fact that violet or
blue purple precedes yellow in Pansies by analyzing the
colors, selecting of course suitable varieties. The writer
has often found a bud pure violet and an expanded flower
of the same plant pure yellow !

The yellow Pansy as it exists today was undoubtedly
once of a purple or violet color, as is clearly shown l)y the
darker face markings almost invariably to be seen and by
the violet spur.

39

Native violets, both wliite and yellow, show vestiges of
the earlier violet color, and the wild, violet colored violets
have no yellow at all I Similarly to Pansy, many varieties
of yellow Iris, Wallflower, etc., may be traced to an older
pur})le or blue, the intermediate colors being reddish-brown
or terra-cotta.

Antirrhinum majus, Snapdragon ; the colors of its differ-
ent varieties can l)e traced from purple, through crimson,
brown, cardinal, terra-cotta, yellow-ln-own to i)ui-e yellow.

Onicus horridiilus, the common yellow thistle, varies from
cream color to pale yellow, while the anthers are a lively
red-purple, and in striking contrast with the other parts of
the flower. The corollas extend below into a long, thread-
like tube of a weak, pearly white color, in fact the funda-
mentil white tis'sue which in the ultimate analysis is the
basis of all color, as not only yellow, but red, purple, and
blue in similar circumstances gradually fade away into this
primitive tissue-white. It is more than probable that this
pale yellow was not the original color, but that the latter
was more like the anthers and the tips of the bracts, which
last in bud are of a pronounced red-purple. The corolla,
once purple, faded to white, from which fundamental color
the yellow is now forming. Another indication that this
yellow is not the i)rimary color is found in the fact that it
does occasional 1}' occur of a [)urple color. The writer has
found a single plant bearing rich, glowing purple flowers in
a field where every other })lant bore the ordinary yellow
])lossoms !

Boletus suhtomentosus, a mushroom, its flesh yellow,
changes innnediately to blue on fracture !

Ixias are of a great variety of color in cultivation, but all
have a dark centre or eye varying in hue from dark pur[)le
to bright crimson, according as they are affected by the
general color of the [)etals. This i)ui'i)le eye indicates that
the original color of the flower was either purple or some
combination of red and blue ; yet there are yellow varieties

40

and certain varieties at first yellow, excei)ting the eye,
change to white and afterwards become crimson, the three
distinct colors being seen on the same raceme, viz., yellow
at the top. white lower down, and the lowest or oldest
flowers crimson ! But in some varieties the crimson shows
in the green bud before any yellow color develo})s and the
flower becomes wholly crimson. Another strain of Ixias
adds red to the yellow, producing scarlet of various tones
and certain varieties of the latter change from scarlet to
crimson ! and finally with age become a })urple.

(TOuUlieria procumbens: The calyx is wholly a waxy
white, like the cylindrical corolla, but the two minute bract-
lets are red I At length the calyx which forms a part of
the berry becomes green, and Anally changes to scarlet !

Lobelia Cardinalis, to an ordinary observer, is simply a
scarlet flower of one hue throughout, but on closer inspec-
tion, one finds the petals and filaments scarlet, anthers dull
blue or blue-l)lack, tip[)ed with a snow-white fringe, pollen
yellow, style greenish ])artly tinged with red, stigma has
two crimson lobes surrounded by a fringe of white hairs,
and calyx green, more or less stained with dark [)uri)lc.
These minutie are best seen by the aid of a magnifying
glass.

Centaurea Cyanus, the corn flower or bachelor's button,
is common in gardens. Select a light colored flower, i)ink
or white ; the scales of the involucre are green with a Iflack
fringe : with a maanif vino- olass find five colors.

CHAPTER VIII.

STRIKING CONTRASTS OF COLOR IN THE SAME FLOWER.

The green calyx is often found to l)c reddeninsi' or pur-
pling at the same time that the corolla iw .some other color,
as white or yellow. This incipient change of color contin-
ues in some species until a clear l)right color is cstal)lislied,
contrasting strongly with the color of the petals. This fact
is of itself sufficient proof that original colors develop in
different ways.

Examples of diversity of color in the different whorls of
the same tlower are numerous, even among our native flora ;
we will refer to a few of these, but give a more extended
list of exotics. It is often the case that the outer whorl of
calyx is so much more developed that the inner is completely
hidden from view, as in Masdevalias.

.Delphinium varief/aium, sepals i)uri)le, petals white.

Do. nudicale, se[)als red, petals yellow.
Polygala saiKjuinea, sepals pur})le and green, })etals yellow.
Heliant]te)num Canadeuiie, sepals tinged with red, [)etals

yellow.
Senecio aureus, tips of green sepiils dull purple, ilowcr

yellow.
Leucothoe racemosa, calyx and ))racts dark red, corolla

white.
Prenanthes alba, involucre purplish, })etals cream (-olor.
Nupliar advena, sepals green, brown within, petals yellow.
Anaphalis mar//arifacea, involucral scales snow white,

true flower is diminutive, and of a yellow color.
Billbergia vittata, flowers indigo-blue, tubular calyx car-
mine.
Iris Kolpahowskiana, falls red and yellow, uprights blue.

42

C/erode)idroti Balfoiiri, calyx inflated puie white, flower

deep crimson. The calyx changes to i)ink with age.

Stamens always pale green.
Phajus Bernaysi, sepals and petals china-white on out-
side and pale yellow within.
Slreilitzia refjixce, a gorgeous flower with large orange

sepals and ultramarine blue petals !
Ahutilon vexiJlarium, cal3'x bright crimson, the partially

protruding petals yellow.
Aechmeas fulgens, flowers scarlet tipi)ed with blue, and

bright scarlet bracts.
Oereus Lemairi, and other species, have pure white petals

and bright yellow sepals.
CerhitJie retorta^ flowers lemon yellow, tipi)ed with deep

purple, floral leaves, several pairs extending down

the stem rich purple.
Do. aspera, flowers red-pur[)le, ti[)i)cd with yellow,

floral leaves blue, tii)ped with yellow !
Fuchia procumbens, deep yellow tube, dark pur})le lobes

and light blue stamens.
BougainviUea, three large pink leaf bracts, enclosing three

inconspicuous narrow linear flowers, reddish outside

and yellow within the lobes
Iris Tingiiana, sepals blue-purple with yellow tips, the

three inner petals are blue-purple.

LEAVES OF BEAUTIFUL COLORS.

Ano'cfoc/nJus, has species with broad orange and green

stripes ; green veined in regular lines, and covered

with a network of gold.
Coleus, varieties innumeral)le, including crimson, crimson

and yellow, purplish-crimson, dark maroon, yelh)w,

tinted rose, etc.
Asjndistra eJatior, has a variety with alternately striped

green and white.

43

Begonia Rex, leaves deep green, banded with a broad
silvery zone, or deep bronze-red centres; spotted
with white, or l)rown-purpie, nerves red, etc.
Caladium, green bU:)tched with white ; carmine-red bor-
dered with white; uniform golden-yellow, etc.
Codid'um, Syn. Croton, green with infinite variegations
chiefly with yellow, which frecpientl}^ changes di-
rectly to a fiery scarlet or crimson, and the lower
surface to purple-brown.
Dracama, green, marbled and banded with various shades

of yellow, and margined and veined with red.
Pandanus Candelabrwn, var. green, with bands of pure

white.
Panicnm variegalum, white strii)ed and pink tinted.
Arnndo donax, green, stri[)ed with white.
Aj)helandi'a leopoldi, green, variegated with narrow,

white feather veins.
Strohilanthes I)//eranius, lilac-})urple with pinnate green

veins.
Many trees in our lawns have variegated leaves, especially
in maples, beeches, etc., and numerous shrubs.

In some of the preceding the leaves are far more attrac-
tive than the flowers.

Coleus displays in its many varieties the two series of
Xanthic and Cyanic coloration. Commencing with a deej)
purple breaking through a green leaf, one series leads
through crimson and pink to white, and the other retaining
the yellow element of the chloro})hyl goes through scarlet
and orange to yellow and finally to white.

Even in a plain green leaf of Coleus one may change to
yellow and another without yellow directly to white. A
blue or purple color is by no means obliged to wait for a
yellow to precede it ! It breaks forth when and where it
pleases ; it is perfectly obvious that many times the yellow
is not at all in the series.

The same remark applies to all rijiening leaves, whether

44

autumnal or of mid-summer; .some jjlants have exclusively
but one series, some only the other, while many, like the
Coleus, combine both kinds.

In Plantago lanceolata the tips of the green sepals, and in
Dandelion the tips of the involucre, l)ecome an inky i)urple.

An example showing the constitutional nature of color is
furnished in a purple variety of Norway maple. In the
seedling plant the first pair of leaves is green, but the stem
and second pair is reddish-pur[)le, and this last color slightly
modified may be seen in the full grown tree, in its winter
buds, leaves, leaf-stalks, peduncles, sepals and style. The
flower therefore, instead of being wholly green as in the
common variety, is of a brick-red color, although the petals
retain their greenish hue. There is no doubt as to the
sequence of color here from the cotyledon leaves of a green
color to the next pair purple ! One may go a step farther
and examine the stem of the yellow fiowered Caltha paluH-
tris just below^ the surface of the ground, and find it colored
a beautiful purple, which color developes before the chloro-
phyl-green !

Another example of the fiower color pervading the whole
plant may be seen in Phytolacca decandra, the pokeweed ;
its peduncles, bracts and petal-like sepals are a pink-purple :
the entire stem becomes red, and the berries change from
sreen through red to black.

CHAPTER IX.

FLOWERS CHANGING FROM GREEN TO WHITE.

MoUuf/o verlici'lJala, is white in every detail, except three
green streaks on the outside of the sei)als.

AlUum, the couiuion onion, is white with a reniainino-
green stripe through the centre of each sepal.

OrnitJiogalum umbeUatitm, is white above, and green with
white margins beneath.

Ei(pko7'bu( marginala, has its upper leaves and bracts
green with snowy white, petal-like margins.

Parnassia Coroliniana. White petals show the basal
cohir in the pretty green veining, and at the tip.

Polygonum arifolium, P. Hydrojjiper, P. acrt^, etc., se})als
change from green to white.

Habenaria tridentata, Petals light green, become white.

Sarracenia Druinmondii, has its trumpet-shaped pitchers
green bek)w, with the upper portion white with
purple veins.

Cj/pripediuyn Lawrenceamnn, green clouded with dull
l)urple, has a white dorsal sepal.

Lapageria alba, is a wax-white, lily-like flower, with a
calyx at first as green as the general foliage. The
flower continues in bloom fresh and fair for six
weeks, during which time the sepals very slowly
fade out to as pure a white as that of the petals !

Angrmcmn, primarily of a green color as in A. sesqai-
pedale, in the course of ten days becomes ivory-
white.

A. mperbiun, greenish-white as to five segments, the lip
becomes pure white.

A. candalwii, olive green, large clawed lip pure white,
its spur nine inches long I

46

Nimtiana acutifolia, green outside, white within.

Cormis florkla, the bracts forniina" an involucre to the
diminutive flowers continue green till fully developed,
say two inches in length, then become pure white.

Coraandra, green at the base of the petals has the upper
portion white, or but fi\intly tinged with green.

(ireen leaves of Panicuni, Pandcnivs, Bef/ojiias. Gerani-
wns, Tradescantias, etc., have varieties with mar-
ginal borders or surface markings of white.

FLOAVERS CHANGING FROM WHITE TO YELLOW% OH DARKElt
HUES TO YELLOW.

Lonkera longijiora, pure white, changes in a day or two
to pure yellow ; also other s})ecies of this genus, of
different colors, as cream, pink, etc., change to
yelloW'.

P/iaseolus, Lima bean, and common garden white varie-
ties, change to yellow.

AngrcEcum citraium, white, changes to citron-yellow.

Odontoglossum Harryanuni, brown and yellow, front lobe
of lip pure white, changes to yellow.

(Joelogyne Dayana, white, becomes tinged with yellow
liy age,

PJatydines glnmaceum, wholly white except minute yel-
low lip.

Sohralia leiicoxantha, white, lip golden-yellow.

Nexilielajphyllum jyaldirum, green, lip white.

Dendrohium thyrsijlorum, white, lip orange-yellow.

Lycaste Skinneri, alba, white, li}) primrose-yellow.

CcMleya citrina, yellow, tip of lip white.

Tliunia Dodgsoniana, white, lip bright yellow.

Bollea LaUndei, bluish-purple, lip golden-yellow.

Mlna lobata, Bud crimson, changes through orange to
pale yellow as it expands.

Cohdea arhorescens, Bud reddish-l)r()wn, through orange
or salmon to a clear yellow.

47

Ahonanda nerifoUu, Bud dark led-purplo, Ijecoiiies u

lemon-yellow.
Viola (ncoloi'. Bud doc[) })uii)lc, occuwiouidly changes to

yellow.
Liliuiu Nepelen,'<e, red-})ui'i)lc, ui)|)ei- half of i)etals yellow.
Tul)erose, Jasnnnc and uiany other white Howers fade to

a yellowish color.

Numerous existing yellow tlowers are known to have been

derived from purple ancestors by the still remaining purple

eye ; examples may be seen in IHbisca>i Trionujn, H. escu-

laiidis, Gos.sf/j}inh/, Coreopsis, Lijsuiiachia, Cidochorlis, etc.

CHAPTER X.

The color of a flower is not generally intlueiiced by that
of the sap: Dandelion, Lactuca, SoncJim, Hieracium, Naba-
lus, etc., 1ku-c white s'lji Imt yellow tlowers: Chelidoniwa
has orange colored sap and yellow flowers ; Sanguinarki,
orange-red sap and white flowers ; Asclejnas, white sap and
l)urple tlowers; Apoci/num, milk-white sap with stems,
petioles, })eduncles and calyx of a reddish color, corolla
white and lose, or wholly piidv.

The orange-red color of the root of celendine gradually
changes up the stem to the exact yellow tint of the flower
at the end.

The change of the same colored sa}) to white in the flower
of Bloodroot seems remarkable, and yet yellow to white is
a common secpience in flowers, and white to puik or purple
is also connnon.

48

POLLEN.

Not only the more conspicuous parts of the tiower, hut
the very pollen-grains have every conceivahle tint ; the
most connnon are cream color, yellow, white, i)urple, ])lue,
green, snuff-brown, and chlorotic or colorless.

Agapanthus, a purple lily, has very blue anthers and
very yellow pollen ; Red Azalea has a black-purple anther
containing white pollen ; Calla Lily has a bright yellow
spadix and snow-white pollen ; Cannabis sativa, anther
green, pollen pure white ; Prenanthes alba, anther black,
pollen pale yellow ; Lobelia cardinalis, anther Ijlue-purple,
pollen sul[)hur-yeliow .

In examining the color of the pollen of one hundred dif-
ferent sjjecies of flowers, the greatest numl)er was found to
be yellow, but nearly as many white, and one-half as many
purple, the others scattering, all bemg in about the same
ratio as found in the colors of the i)etals, though often
varying from them.

We will merely allude to other minute organs having dif-
ferent colors developing in different ways, as hairs, glands,
etc., viz. : The hairs growing on the lower side of the green
leaf of Cineraria and other plants are often pur[)le or red,
and the white glandular hairs on Azalea viscosa have little
red knobs on the ends !

While green is generally more primar}^ than yellow, it is
not always the original or most primitive. Any color may
precede and be entirely independent of chlorophyl-green, as
may be seen in examining roots, bulbs, root-stalks and
sprouting stems, also fungi, all of Avhich have colors arising
directly from the structural or chlorotic white of the em-
bryo, without any special order of development.

Familiar exami)les : Beet : roots, stems, etc., red ; Carrot,
reddish yellow ; Parsnips, from cream color to pale yellow ;
Turnips, with purple to[)s and yellow turnii)s with puri)le

49

bases ! OnioJis, of all colors ; Radish, pure crimson ; Mede-
ola, Indian Cucumber-root, white; Coptis, Goldthread,
bright yellow, etc., etc.

Likewise many petals show color in their earliest stages,
especially such as are well protected with sepals, as the Red
Rhododendron, etc., without any development of chlorophyl.

DOUBLE FLOWERS.

Whenever stamens change to petals as in doul)lc flowers,
whatever the color of the anther previously, the new growth
caused by the unrolling of the anther and tilament which
may be of different colors, innnediately takes a new de-
parture in color. Frequently a i)art of the original anther
with its original color will remain attached to the incipient
petal, as a blue anther cohering to a crimson })etal in Tulips.
The prominent green stamens of Nigella Dainascena change
directly to a clear blue color, etc., etc.

Why does the color change so completely unless there be
special needs for special organs?

Double, so called, English Anemone, has a tuft of green
leaves in the centre of the flower, why do these turn white
like the petals when they get their growth, or similar leaves
of pink flowering Almond change to pink, or l)oth leaves
and stamens of white Spirtea become white?

PIGMENTS.

Yellow, green or orange pigments are found in variously
formed granules in the cells of the tissue: blue, violet, and
rose-red are dissolved in the cell sap. Chlorophyl, con-
tained in the green granule, can l)c abstracted by the use
of alcohol, and by other chemicals separated into two sub-
stances, one of a bluish-green color, and the other a yellow
which has been termed Xanthophyl. There is also another
yellow substance found in the green parts of plants known
as etiolin. Xanthophyl and etiolin may be derivatives of

50

chlorophyl l)y oxidation or otherwise. The red eoU)ring
uiiitter has been named erythrophyl.

There are at least two kinds of white color, one a })ignient
named leucophyl, the other caused by the absence of all
color and generally referred to as chlorotic ; this kind of
white is only an o])tical effect of structure, but the two
kinds grade into each other without perceptible dividing line.

Different })igments, as those fixed in the plastids, and
others free in the cell sap, often exist at the same time, one
kind then naturally modifying the hue of the other. With
such a condition, viz., of different elements associated at
the same time in the same flower, one can easily conceive
that a plant widely distributed might under different condi-
tions in one case develop the purple element which is free
in the cell sap, and in another, the fixed yellow xanthophyl,
or erythrophyl substance, so that the difficulty of assuming
that the colors of different species of the same genus were
all derived from one primary color of one original ancestor
would be eliminated.

These pigments are found mainly near the surface of the
flower petal, just beneath the cuticle, sometimes in the skin
itself. In a white Crocus variegated with purple, if the
transparent cuticle is peeled off, it carries the purple color
with it, leaving the remaining tissue white ; not infrequently
however, the pigment extends quite through the tissue from
side to side, continuing even down the peduncle and stem,
as in Arethusa and certain Begonias. The yellow Crocus is
tinged quite through the entire tissue, also Buttercup, La-
burnum, purple Iris and red Pa^ony, Various kinds of
acids are found free in the cell sap, or in cond)ination with
alkalies, and the varying proportions of these elements play
an important part in estaldishing the colors of flowers ; the
effect of these will l^e described later.

As hinted above, there is no good reason for supposing
that there was an original color in any genus from which
all the colors of the other species were derived. There was

51

a time doubtless when plants did not bear the brijjht colored
flowers of the present day ; the}^ were first apetalous and
probal)ly as unsiahtly as our j)resent pond weeds ; then
long before they had differentiated into the present species
or had even beoun to develop petals, there may have been a
marked difference in the colors of the essential organs,
scales or involucres, in different plants widely dis[)ersed, as
is seen at the present day in our apetalous flowers. Take
for an illustrative example the oaks, and suppose they
should And it for their advantage to have bright colored
envelopes for their naked flowers. We find them already
varying much in the color of their leaves, scales and
stamens; compare the black oak, white oak, live oak, etc.,
the leaves of one unfolding green, another brick red, another
a pinkish purple ; the anthers of white oak simply green,
tending to become yellowish, and those of scrub oak a
bright scarlet. Would it not seem entirely natural for each
species to develop its own special color? and if it should,
then the yellow and the scarlet colors which might ai)pear
would be contemporaneous, one as primitive as the other.

Iris has yellow, also purple or violet colored species simi-
lar to the Pansy, which has both colors and commonly no
red intervening. Our New England Iris versicolor and /.
prismatica are violet with some green, and the more south-
ern P^uropean and Asiatic species are yellow.

Suppose one should enquire, Which of the two is the
earlier or primitive color? Probably Grant Allen would
say, the yellow of course. We would say, however, that it
does not follow as a matter of course for these reasons : in
our native Iris, the broad green claws of the three outer
segments are prettily veined with purple, while the broader
blades or tops of the falls are pure, rich violet. We have
shown hereinbefore how a purple color often naturally
breaks through the chlorophyl green, in leaf or flower as in
Cypripedium, Coleus, etc., or may even precede the chloro-
phyl green itself, as in purple foliage shrubs, trees, etc.

52

We have also shown that it is just as natural under suitable
conditions for the orreen color to dissolve into a clear
yellow.

In our example, the violet exists in the cell sap, and the
chloroplastids are tinged with green at the same time. It is
a question simply as to which shall prevail, and that will
depend upon chemical conditions at present not sufficiently
understood. In our Iris the violet prevails.

The green upper portion of the claw fades away, leaving
a few white streaks extending into the violet blade ; this is
a common sequence : green to white ; with age that portion
of the sepal at the height of the declined style becomes
somewhat yellowish, that is, a very small })art of the long-
green claw has transmuted its chlorophyl to Xanthophyl,
and that is all the yellow there is in the flower, the inner
segments or petals being of the same green and purple,
with faint whitish streaks only.

The yellow species upon examination will be found to
have always more or less of purple-brown on the claw^*, this
color being a purple or violet modified 1)y the yellow.
Some species are wholly of this mixed purple-browni, as
shown in Iris fulva. It appears, therefore, that the two
pigments really exist in all the species, and one is just as
likely to be as prominent as the other, according as climate
or other circumstances favor.

The reason why yellow is more conunon than red or l)lue
I apprehend, is because the ordinary yellow is dependent
upon a previous green generally present in the flo^ver bud,
and when this recedes its yellow constituent is apt to
remain. I have not infrequently seen a white, lavender, or
red-purple flower, when the petals first })eep out of the
calyx, to be more or less yellow temporarily.

In observing the colors of the common garden Sweet Pea
one frequently remarks a variegation in the same flower :
the banner being of one hue an<l the wings and petals of
another. The banner is the outer petal which enfolds and

53

protects the others in the liud, and consequently is at first
somewhat green in the early state. As a consequence while
this banner is never really yolhnv, yet the chlorophyl green
in dissolving often leaves a trace of it sufficient to affect
the subsequent colors, which from this cause are frequently
of the xanthic series, while those of the wings and keel are
of the cyanic series. I have never (observed them to be in
the reverse order ! Of course the banner may be pink and
the lower parts white, or the fiower wholly white, but when
the xanthic colors do appear they are invariably up[)crmost.

Flowers for the most })art have their peculiar, stable and
uniform colors, as in the wild rose, buttercup, strawberry,
etc. ; the number in which important variations have oc-
curred is, however, considerable ; and while different species
do not always have different colors, yet if a plant is found
with an entirely different hue, the probability is that if it
persists and nudti plies that it will also show^ other character-
istics, as differences in sha})e of leaf, or habit, sufficient to
entitle it to the rank of variety or incipient species. — But
why one tlower is yellow and another blue is a difficult
question to answer categorically. One might as well en-
quire why one plant produces gum, another sugar, a third
oil, others acids or alkalies, and a hundred other special
products. Of course one does nf)t expect to find as impor-
tant changes in the physical conditions and chemical projjcr-
ties of the plant to produce the ordinary color })igments as
the former products.

Green is par excellence the color of growth and vegetation,
whether in leaf, flower or fruit; hence while growing and
until they reach their normal size, flower buds and fruits
alike are green, and naturally from this chlorophyl-green,
with additions, or from its derivatives, or from pigments
entirely independent of it, develop finally the numberless
hues of flowers and fruits ; for this reason we have in this
essay given green a prominent [)lace as a primitive color,
while not overlooking the fact that tlu^ greater part of all

54

colors under certain circumstances might precede this green
and also that more primitive than all is the fundamental
sh-uctural or chlorotic white.

CONCLUSIONS.

We tind that among the Ranunculacea', yellow is not by
any means the prevailing color. In Clematis ydJoir is
comparatively a rare color.

Very many existing yellows are known to have been pre-
ceded by entirely different colors. Brown-purples and sim-
ilar hues formed of green and red are earlier than yellows.

Simple or primary yellow does not regularly change
through white to led or purple ; an exceedingly common
sequence being yellow to orange, vermilion or scarlet, or to
brown ; or through green to l)lue, or directly to crimson.
Even in the Buttercup, there is a change of yellow^ to scarlet !
Not infrequently green and [)urple, green and brown, green
and red, green and white, green and violet, etc., ai)pear
before any yellow color develops. One w^ith a trained eye
for color can scarcely look at a flower without detecting a
difference in the development of its hues ; the anthers per-
haps diverging into one series of color, the petals into
another, the pollen often different in hue from the anther
containing it, and possibly the sepals different from either,
as in Lobelia Cardinalis ; it frequently occurs that the very
hairs ((re of a different color from the stem or leaf from
which they grow. Size, shape, or construction, as a rule,
has no effect on color, neither simplicity nor complexity.

Nature is highly democratic in her feelings; she bestows
the royal purple or cerulean l)lue as freely upon Clematis,
Hcjiatica and Anemone of the simplest family as u})on
Aster, Hyacinth, and Gentian of the highest families, and,
conversely, she distributes the golden yellow on the wdiolc
in as great a pro])ortion among the higher as among the
simpler species.

HONEY GUIDES.

CHAPTER XI.

Flowers u.siuilly have hut a .sinirle, uniform color, de-
pending upon the species ; tliere uie, however, many which
are variegated with more or le»^s reguUir spots, streaks, dots,
and "eyes," of a different color, usually in strong contrast
with the predominating hue.

When these additional colors are contined to a single
petal, they generally point to the spur of the Hower if it
has one, or to the nectary or depressit)n containing the
sweet secretions; they are known as "honey guides," or
" pathfinders, " for the reason that they have been supposed
of use in guiding insects to the nectar.

Examples of the guide-markings may be seen in the
following flowers :

Mi)i/iilus rinyens of a lavender-blue, has a yellow spot at
the throat.

Pontederia cordala, ))ickerel-weed, violet-j)ur[)le, has a
two-lobed yellow dot on the middle of the uj)per
three-lobed petal.

Lef>pedeza poly stack ija, cream-color, has a [)uri)lc si)ot on
the standard.

Sisyrinchium. ((U(/ii,s(iJ'oUi(m, blue, h;is a yellow eye.

Solanum dulcamara, violet-purple, has ten round green
dots, two to each {)etal, surrounding the column of
stamens.

In garden Balsam of many vai'ieties, the honey-guide is
a meie yellow dot of the size of a [)in liead, and is found
on one of the sei)als I

Tropd'olum or nasturtium, yellow or orange, has })urple-
maroon veins or wider stri})es, mostly on the two upper
petals.

0()

Of a siiiiilar nature ii^ the scarlet-maroon eye of yellow
Coreopsis, the purple centres of white Phlox, Althiea, Hol-
lyhock, etc., the criaison zone on a white ground in some
varieties of Sweet William, the yellow "eye" of scarlet
Tulip, the large black spots on the petals of oriental I^oppy,
the different colored lips of Orchids, and the golden bands,
and red, glandular dots of LUiiiin auratiun, etc., etc.

Were these different colored markings a part of the
original design of the.liower, or have they been added to a
plain self color, and if so, by whom and why?

Christian Conrad Sprengel's attention was called to this
subject in 17<S<S. He first noticed the yellow ring round the
mouth of the corolla (jf Myomtls palustrls, L., and "he con-
ceived the idea that this might serve to guide insects on
their way to the honey. On examining other flowers he
found colored dots and lines and other figures at the en-
trance to the nectary or pointing towards it, and was accord-
ingly confirmed in his idea of [)ath finders or honey guides."

W. E. Shuckard in his book of British Bees says :
"Flowers, within themselves, indicate to the l^ees visiting
them the presence of Nectaria, by spots colored differently
from the petals. The honey marks surely guide them.'"

Grant Allen writes as follows regarding these honey
guides: "We can hardly resist the influence that they have
been cieveloped to guide the bees to the proper ])lace for
securing the nectar."

Prof. iNl. I. Xewbigin in a recent work says: "Their
meaning and origin are still very doubtful.'"

Alice Lounsberry, in her "Guide to the Wild Flowers,"
refers to the "Road that is plainly marked out for the bee,
by a deep, rich veining"" to a gland of nectar.

G. Henslow remarks: "It has l)een observed that the
'spots' are more persistent than the l)asc color of the flower.
W^henever they occur as auides or })athfindcrs, which have
been determined by insects, they have become hereditary."

John H. Lovell in the American Naturalist for June,

57

1899, writes as follows : "In a species of Dendrobium the
sepals and petals are white, ti[)i)ed with })Lirple, and the li^)
is bright orange with two crimson s[)()ts. This is probably
due to the marked tendency of the si)ecies to variation."

C. Darwin's reference to iionoy guides will I)e found
farther on. Of these se\'en writers of repute, oidy one
hazards the (jpinion that they "have l)een determined by
insects," but not a hint is vouchsafed of the manner in
which they were determincnl, nor by what kind of insects.
Neither have I ever seen or heard of any explanation re-
garding their origin ; I shall therefore offer the following
as an original elucidation of the subject.

For our first example in illustration, let us take a connnou
yellow garden Tropicolum, otherwise Nasturtium.*

The plant is either climbing or dwarf, with round peltate
leaves, and the long-spurred flower has many varieties. We

'The pen aud ink sketches iire l)y Joseph E. Lander?

58

will select as best adapted for our })ur})ose a Hower of a
yellow color with a longitudinal reddish-brown stripe on
each of the upper petals. These are the honey guides and
they in this example really point direc^tly to the nectary.
The (juestion to be solved is how they got there.

Did the Creator design them expressly for the benetit of
nectar-loving insects, or did the insects })aint these ])etals
themselves with tlieir own saliva, same as a grasshopper
will stain your fingers if you hold him for a moment? It is
of course remarkable that insects should ])e so cared for as
to have a s})ecial sign-board directing them by the shortest
route to the sweets they so much crave !

If this }>articular variety was the only one to l)e found,
our task would be harder, but as we examine the numerous
sorts, we will be very likely to run across a dark colored
flower with no honey guides, and whose color is exactly the
chocolate or Indian-red of the markings on our yellow
example. In the yihice where the honey guides ought to
be, the color is of the same as the general hue, only a trifle
richer and deeper. This richness of color is (X'casioned hy
the irritating influences of the bees in traversing the same
route to and from the nectary, thus stimulating the flower
to send more of its peculiar })igment to this })oint, same as
a little friction or a pinch will bring the blood to the cheek
and cause a rosy tint.

The next step in the })rocess is to change the general
color of the flower to yellow. How this is effected by
Nature it is unnecessary to ex})lain just now ; suffice it to
say that the fading of a color to a lighter tint is one of the
most common occurrences in color-change ; we have only
to look around and ol)serve the yellow roses fading, some
to a clear white on the same shrub, and the pink ones also
fading out, and scores of other flowers blanching on their
own stems ; blue Houstonia and Myosotis, the pink English
daisy and apple-blossom, changing to white, etc., but in
our Tropjeolum the flower never ))ecomes })erfectly white ;

59

it docs, however, change to much lighter tints, some as pale
as cream-c()h)i-. In our search every gradation is found,
l)ut in every case tiie deeper and therefore more permanent
shade of the honey guides remains, only somewhat nioditied
in color. Thus we get l)ack to our yeUow examj)le with its
Indian-red honey guides, the reu)aining vestiges of its
original color I In this same example there is very likely
to he found smaller spots on the three lower petals ; these
are less marked for the reason that the smaller insects,
which are careless and don't stop to read the clearly letteied
guide-board of the upper petals, foolishly alight below,
where they are del)arred from getting at the nectar by a
sort of cheval-de-frtse^ or erect, fringe barrier ; but the
silly insects stimulate just the same, though in such less
degree that the three spots not infrequentl}' wholly disap-
pear, while the two upper, more thoroughly made, will
persist. We have in these lower spots a hint that honey
guides do not al\va3^s guide !

Example 2. — We will brietiy describe another example.

One vari(!ty of the slender S{)anish Iris is of a beautiful
itlu(\ with a golden-yellow spot on each of its three outer
segments. It is more wonderful to behold than the Nastur-
tium, l)ut, as we have fully descrii)ed above the principle of
honey-guide decoration, all its mystery disappears.

As in Nasturtium, this ilower has many varieties. The
typical tlower is entirely yellow, with no markings (except a
little more of a Naples yellow tint in the place of a honey
guide, but practically no guide at all ; it might be added in
this connection that the bees get along just as well in the
yellow flower with no guide, as the}' do in the blue flower
with the brightest possible sign to arrest their attention : it
all de[iends upon habit !

Rut to return to our subject. The general yellow in this
case. fades completely to white, except the markings, which
remain as in Tropseolum ; hence there is a white variety
with a yellow honey guide ; then begins a tinge of blue on

fiO

sonic of the white varieties, which oradually deepens and
spreads u]) to the very limits of the uolden yeUow spot,
without encroaching one hair upon its territory. Notice
two |)()ints : the tiower change is from white to blue, just
as is seen in Pansy, with not a hit of red intervening; also
that sometimes the change is so innnediate that it is from
yellow to hhie, no wdiite even being noticed.

Anything that arrests the attention may be a valuable
guide for certain purposes ; the stamens of the fragrant,
fuU-chistered, little white Rosa multitlora are at tirst of a
bright yellow, by the second day or I)efore they turn a dark
brown, so that the yellow and the l)rowm centres are quite
distinct in the same cluster. The bees always take the
yellow centres and pass the others by.

Example 3. — In violet-)iurple Ponlejhria ^ pickerel-weed,
the bum))le-bee always Hies directly towards the t\vo-loI)ed
yellow dot on the upper petal, against which he presses his
head as he sips the nectar.

In this instance it is the insect's head and not his feet that
causes the irritation, preserving most likely the original
color of the flower, the dark i)urple l)eing the more recent.
It is quite possilile also that buttertlies are concerned in
this particular marking, for the })roniinent little feelers
between their eyes touch at exactly this place as they
extract the nectar !

Example 4. — In GalantJius, the snow-drop, and Leuro-
iu/H, the snow-Hake, we have examples of green colored
honey guides; in the former of these flowers which are
both white, the three short inner segments are green on the
inside where the insects enter ; m Leucoinn) the bee clings
tirst to the drooping tips of the sei)als which are green
colored, before he turns up into the Hower, These green
tips, all the rest of the flower being white, were caused by
the bee in the same manner that the yellow spot of Iris
and the Indian-red of Tropteolum were i)roduced, but the
original color of Leucoium was green I

(U

Example 5. — The widely .scattered dot« on LUium aura-
luiH were not accounted for by the writer until a huuiniing-
hird moth was seen to hover over the flower, and suck in
succession the nectar from each red, glandular vesicle !

Plxample 6. — TrUlium eri/throcarpum, painted trillium, is
a beautiful white flower, painted with a crimson eye with
fine radiating lines. The eye is also of the nature of a
honey guide. It is the remnant of a former crimson flower
and preserved to us by the insects !

Example 7. — Viola trirohn-. A yellow pansy has three
dark maroon or brown-purple sjjots ; the two ui)per smaller,
the lower large and two lobed. A violet pansy has only a
very small yellow eye at the centre.

The ordinary yellow variety with the maroon si)ots was
undoubtedly derived from the violet variety.

Example 8. — Eschscholtzia lias an orange, self color, but
its lemon yellow variety has four orange spots correspond-
ing to those on oriental poppy. The lemon yellow is a
recent color, the flower retaining the orange markings of
the original.

P^xample 5). — Gloxinia, Digitalis and Catalpa : The orig-
inal purple or crimson fades to white on the outside of the

62

tube, retaining the color naturally longer within, hut it
finally breaks up into numerous })urj)le clots. The two addi-
tional streaks of yellow in Catalpa may be a more recent
color [)roduced when the flower became bi-lateral and the
ridges formed, necessitating the insect to take that course.

Jmpatiens fulva is similarly marked with purple dots
equally distributed over all the lobes.

Example 10. — In the preceding examples we have shown
how a bee can assist Nature in variegating a flower with
another C()h)r. We are al)le to demonstrate farther that a
bee not only assists, but actually produces color, and that
the same bee may be instrumental in developing three or
four distinct colors in the same flower ! We will endeavor

63

to illustrate this idea by talgng for an example the common
yellow variety of garden Tulip.* The purest and most orig-
inal specimens are yellow in oxory detail : perianth, stamens,
stigma and also the very grains of pollen. From this lemon-
yellow tulip are derived a great variety of charming hues
and many pretty variegations. We shall not undertake
here to explain the chemistry of these color changes, hut
we will point out the role of the insects in this magical
transformation.

Variegation 1. — As one of the ways colors run, and not
l)y any means a necessary sequence, as this very example
will clearly prove, we find individuals of this flower which
have changed to pure white except a yellow area at the
bottom of the chalice which still retains the original color,
so that we have a white flower with a yellow eye. ^^'hy
does this yellow eye remain, is it of the nature of a honey
guide? If so, it certainly does not guide, for there is never
a drop of nectar in the cup ; it is rather a delusion and a
snare ! In watching the bees, one flnds that they do not
alight on the petals, nor do they force an entrance before
the bud has widely expanded, but when it is invitingly
open, the bee is able to fly directly into the opening and
alight upon the stigma or the anthers which alone afford the
nectar. For this reason there are no guiding honey guides.
The insect does not always retain his foothold, and the
inner segments of the perianth being erect, and smooth as
glass, with a slight concave bend near the middle, if he
trusts to these he is sure to fall to the bottom, from which
it is no easy matter to escai)e. Not only bees l)ut other
small insects, clamber up to fall again, and are finally
exhausted and smothered in the pollen dust accumulated at
the bottom. Their struggles for life preserve to us the
golden eye, a reninant of the original hue ! I have found
in one flower a bee, bug and ant. This is a cold, hard-
hearted flower and dangerous to insects.

* The illufitration is a wilted Tulip, to show markiug at centre.

64

Variegation 2. — There is a slight tendency to blue, mani-
fested in the anthers of the same original type, which by
the stimulus of constant visits of bees is increased and the
color deepened, so that besides the original yellow and the
white, one may occasionally see an anther which has turned
blue, and the darker the flower, that is, if it lie a scarlet
for instance, the surer one is to lind a l)lue anther, and not
"only this, but the yellow eye changes to blue, and the same
bees w^hich were responsible for the yellow eye are also
responsil)le in a measure for the dark cobalt-l)lue of T. Ges-
neriana and for the black pollen dust changed from golden
3'ellow.

There is a very attractive variety of tuh'p known as the
Kaiser Crown, of yellow and scarlet colors ; its peculiarity
is in the two ways of variegation. The scarlet which suf-
fuses the yellow foundation color runs on the outside, from
base towards the tip ; on the contrary, that on the inside is
the reverse, that is to say, the scarlet color extends hori-
zontally on each petal, forming a central scarlet zone in the
yellow tlower-cup. The former is the natural method ; the
latter is the result of insect agency, and is,

Variegation 3. — The same l)ee while giving his undivided
attention to the anthers, owing to the proximity of the
erect petals, is ai)t, in going the rounds of the stamens, to
rub his back against them, wnth the result that just where
it touches is left a more or less distinct scarlet band within
the petals at the exact height of the anthers I This feature
being noticed by the Dutch tlorists was doul)tless seized
upon and hence the beautiful variety.

This singular zonal marking cannot be seen in every
yellow flower. It is more apt to l)e found in one which
has a tendenc}^ to a scarlet or red color, seen perhaps on
the outside, and the stimulus of the insect awakens it
within. In like manner and b}^ the same agency, the writer
has seen a wdiite tulip with a crimson zone on the inside,
the "eye" still remaining yellow and the anthers blue.

65

G6

makino^ four distinct colors ! Q. E. D. — These last colors
are accounted for hy the disappearance of the yellow pig-
ment, which leaves a white basal color and changes scarlet
to crimson. The yellow eye of a white tulip is apt final I3'
to fade to white. I have also observed a similar red zone
in a white Ron)an Anemone.

Prof. M. I. Newbigiu mentions what appears to be a
mystery to him concerning honey guides, A'iz. : the fact that
they occur sometimes where there are no nectaries. No
examples are given, but the professor might have had in
mind a flower like the Oriental Poppy.

Example 11. — This scarlet poppy* is one of the largest
of flowers known, l^eing quite seven inches in diameter, and
on each of its four broad, flaming petals, near the l)ase,
there is a black spot one inch in diameter. As Prof. N.
remarks, there is not the slightest evidence of any nectar
anywhere on the petal and apparently none in any other
part of the flower. What then are the spots for, and how
were they produced ?

Of course, we have no record of the history of the flower
and of what colors it flaunted long ago in the garden of
P^den, so one must theorize somewhat, and our theory is
just what has been stated regarding the tulip ; evidently the
flower has changed from its original color which was some
shade of purple, and the bees or some other Eastern insect
visited the flower just as they do now for the nectar that
the anthers afford, precisely as the bees visit the anthers of
the rose in these days. As they revolved around among
the numerous stamens of the same inky hue as the spots, in
clinging to the outermost, their backs rubbed against the
walls of the petals at the precise place of the spots, which
is central on the petal and slightly depressed. The sta-
mens are weak and inclined to drooj), so that the weight of
the bee throws him, or if he loses his hold, he falls on to

*Tlie illustration, p. 65, represents a small variety of Oriental Poppy.

67

the (lci)ressions. This constant irritation of the cuticle at
this one point deepened the color and retained it, wlien
every other part of the jietal was transmuted by chemical
agency to a Hame-like hue I Of course the prominent,
black honey guides are of no value as real guides, they
serve to decorate the petal and that is all ; and this is all
that half of the spots or guides are good for I

They may, however, l)c of some utility for a purpose to
be referred to later.

Example 12. — Llriodendron Tuhpifera, Tulip-tree, is a
stately tree, bearing cup-shaped, green flowers encircled on
the inside with a remarkably rich orange band, which color
strikes through to the outside.

The flowei" is two inches across, with long stamens reach-
ing nearly to the U)\) of the six upright petals, and its three
sepals, (juite similar in size and shape, but lacking the
orange color, become strongly declined.

The general appearance of the tlower is that of a tulip,
and the orange band recalls a similar band in the garden
variety of tulip, before referred to, though differing in tint.
When I first examined the tlower 1 innnediately remarked
that the orange band was low down in the tlower, while
according to my theory it ought to be directly opjjosite the
anthers, so that the bee, as in the case of the tulij) and
[)oppy, would come directly in contact with the spot or
marking on the petal ; this clearly was not the case in
Liriodendron, and for a moment I was j)uzzled. I was
able, however, shortly to discover the reason of this differ-
ence. One cannot examine the actions of insects quite so
handily on a tall tree as he might wish, but happily there
were a few low branches and I was not obliged to clinib it.
Greatly to my surprise the bees flew right past the anthers
into the space between the stamens and the orange band,
and instead of facing the anthers, they faced the petal and
confined themselves strictly to the orange zone, not going
above or below it ! All the surface of this zone they dili-

G8

gently licked over, making a complete circuit of the corolla.
The position of a nectary and at the same time of a genuine
honey guide so much above the bottom of the chalice is
very unusual indeed ; it is the only example I have ever
met.

There was nothing discerned by the naked eye to attract
the l)ees but the color ; with a lens however the orange band
was seen to be minutely and obscurely punctate, and the
pollen falling from the extrorse anthers upon it, mingling
with the secretions, formed a delicate ambrosial nectar.
This nectar originally was the lirst thing to attract the bees,
then their stimulating influences evoked the color ! I have
had suspicions all along my researches, that insects did to
some extent awaken a new color ; they were aroused in the
Kaiser Crown tulip, where it was the only scarlet color on
the inside, and produced by bees, but the pigment was in
the "blood" of this flower, for it showed naturally on the
outside, but in Liriodendron there is not a particle of
orange color elsewhere !

This pigment no doubt exists in the petals undeveloped,
for with age the green color becomes somewhat yellow ;
but this elegant band of gold on a green ground would
never have ap})eared without the aid of insects, and in this
instance, therefore, I should say that it was not a vestige of
the past, but a new creation !

Example 13. — Sisy^Hnchium auguslifolium and >S'. anceps
are two common New England flowers, small l)ut interest-
ing. They are blue flowers having a yellow centre. Each
blue petal has a diminutive yellow spot at the base, and
these six together form a small star-shaped eye ; the base of
the style to. the exact height of the eye is also yellow, but
higher up is l)lue. This flower was undoubtedly in the
remote })ast of an entirely yellow color, perianth, stamens
and style. Possibly it emigrated from the tropics, where
its congeners arc still mostly yellow. It may have taken
thousands of vears to reach New England, and the chano-e

69

of climate pr()l)al)ly changed its physical character sutH-
cientlj to alter the relative proportions of acids and alka-
lies, or even to develop a new element requisite for pro-
ducing the blue pigment. At all events, the tell-tale eye
stamps it as of foreign parentage, and we know from the
foregoing illustrations why the eye is yellow, and why one-
half of the style is also yellow and the other half blue !

Example 14. — The little Lobelia of a blue color, fre<iuent-
ly seen in hanging baskets, has for honey guides a pair of
diminutive white streaks at the throat, slightly diverging.
It is visited by a little bee a third of an inch long, and just
strong enough to force open the palate and enter for one-half
the length of its body, dragging its hind legs exactly over
the guides. These divergent little streaks therefore are
not exactly footprints, but im[)ressions of the insect's legs
as it lies flat and outstretched si[)ping the nectar !

Charles Darwin points out "a striking case of correlation
of color with a special organ," viz.: ''In many Pelargo-
niums the two upper [)etals in the central flower of the truss
often lose their patches of darker color: and when this
occurs, the adherent nectary is (juite aborted, the central
flower thus becoming })eloric or regular," etc. Well, that
is precisely what one would expect, no nectary no honey-
guide !

Darwin was a very busy man and he would not have l)een
long in flnding out all about the nature of honey guides if
he had only turned his attention to that sul)ject, but one
would judge from the quotation that he had not done so.

Grant Allen in The Strand Magazine for April, 1899,
referring to white flowering plants, remarks as follows: "A
curious coincidence is, that none of them have any streaks,
spots, or lines on their petals. The reason is simple.
Such streaks or lines are ahvays^ honey guides to lead the
insect straight to the nectary " !

Anything that arrests attention may be useful as a guide
for insects. A large [)ro})ortion of the markings on flowers,

70

as we have seen, do not guide to the nectary. Some do so
point, ])ut of these many are in the tube, and of no vahie
when the insect has got so far towards his object. Never-
theless, while many of the spots, dots, bands, etc., are not
guides to the nectary, they serve as signs or advertisements
to insects that such flowers are especially nectariferous.
Probably there is no flower with these markings but that
has nectar somewhere in the blossom. They are always
confined to those parts only of the flower which come in
contact with insects. It is the contact which causes them.

Other useful guides are found in the rapid change of
colors, Weigelia rosea opens white, and changes to deep
rose-purple ; after a time there are but few white flowers
and many purple on the shrub. The honey bees visit only
the white ones.

A bumble-bee will fly among a bed of Columbines, per-
haps all white ; he goes close to the flowers, looks into this
or that, and passes on to others, finally entering some. If
the ones he visits are examined, they will all ]>e found to
have bright lemon-yellow anthers, while the ones he rejects,
though looking just as fresh, are not so, all their anthers
have turned to black. Even a change of color in the spots
themselves, to which we have previously referred, serves a
like purpose, as it indicates that the flower has passed its
virgin freshness, and with it its nectar has dried up.

Real double flowers have no insect visitors, partly double
do, and the markings may continue on these for a time by
heredity.

The honey guides on Polygonatum, Solomon's Seal, and
Lucoium are green ; on Phlox Drummoudi and Cineraria,
white at the centre frequently, while Convolvulus sepium
has white, radiating streaks ; they are Indian-red in Tropte-
olum ; yellow in tulij), iris, freezia, pontederia, etc ; pur[)le
or violet in pansy and white violet, etc. ; cobalt blue in
Tulipa Gesneriana ; and violet-black in oriental poppy.

In Rhododendron, Azalea, Geranium, etc., the special

71

niiirkings ;irc on the upper petals, for the reason that the
stamens on that side are shorter and detiexed, also for the
reason that the nectary which the insects themselves have
made are on that side, and chiefly because the insect alifjhts
there.

In Weif/elia rosea, Gladiolus, orchids, etc., the marks
are on the lower side of the tube or lip, as being the more
natural resting place, when the stamens do not interfere, if
the flowers are declined or horizontally disposed.

Upright flowers, as tulips and poppies, have similar
markings on each petal, as the insects have equal access
from any point.

When the petals are rather narrow as in Freezia and
some orchids, the markings extend over to the adjoining
petals on each side, probably because the spread of the
visiting insects' wings comes in contact with these side
petals, or the insect does not confine, himself to narrow
limits ; in Rhododendron, the corolla being gamopetalous,
the numerous dots extend over three lobes.

In addition to the exam[)les noted of honey guides or
special markings, are Melampyrum, Calo})ogon, Pogonia,
Lysimachia, Aesculus, Kibes aurea, Oxalis, Gloxinea,
Apocynum, Veronica, Coreopsis, Hibiscus, Gossypium,
Calochortis, Digitalis, Cladrastis, Laburnum, C'atalpa, Cy-
clamen, Baptisia, French Marygold, etc.

Generally speaking, special markings like those described
above, indicate that the flower has changed color. The
markings are usually relics of the previous color of the
flower, but not always so, as they are often modified l)y the
new pigment, and sometimes their hue is completely
changed. In a comparativel}' few instances the markings
appear to be recent. They are always the result of actual
contact with the insects at these precise spots.

A few writers have noted the fact, without any explana-
tion, however, that night blooming flowers have no honey
guides or other special markings.

72

The author of "Field, Fore.st and Wayside Flowers," in
describing the difference between day and night blossoms,
says of the red campion. Lychnis githago, a day bloomer:
"A few clearly-drawn, dark lines, running from the edge
of the blossom to its centre, are a floral signal-code telling
the butterflies where the nectar which they seek is stored,"'
etc.

The evening-lychnis of a white color, "has no lines to
indicate the whereabouts of its nectar, for these would be
undistinguishable in the dark, and therefore useless," etc.
"But a very small object, if it be white, can l^e seen in the
darkest hours of a moonless night."

This of course is not a very satisfactory reason why night
bloomers have no dark markings. We are not altogether
certain that a winged insect which could see a white flower
by twilight, might not notice a red or purple spot on the
petals ; certainly there would be a sharp contrast of light
and shade. But we think this lack of guide markings can
easily be accounted for in another manner.

We have shown that they are incident to a general change
of color, usually of a dark tint to a lighter one, commonly
flrst to white. Among night blooming flowers there are no
dark colors, no red, purple or blue. If dame Nature ever
produced them, she long ago corrected that mistake for, at
present, they are only white, pale yellow, or pink. Such
flowers as Azalea viscosa and Oenothera biennis have
adapted themselves to night shadows and their colors, white
and yellow, are the very best for nocturnal insects, there-
fore they never change, and there can never be any red or
purple spots, simply because there are no flowers of these
colors !

There is, however, one twilight bloomer at least, and
there may be others, having guide markings, I refer to
Convolvulus sepium. It is a pink flower, funnel form, of
the size and appearance of a Morning Glory. The wide
tu])e is white, with Ave white rays extending up through

73

the centre of each lobe to the border of tlie liml), so that
the effect is tliat of a tive-rayecl star on a pink ground.
The white star is tlie honey guide, its Avhite color being the
oi-iginal one and the pink more recent. The flower ))U>oms
in the morning twilight, long l)efore the roses, and lasts
but for a day. The white guide markings were produced
when the flower became pink, and are very appro})riate and
suitable for twilights. There are varieties of the flower
wholly white.

Besides the crepuscular and nocturnal bloomers, there are
diurnal bloomers which keep open house all night. Privet
is visited by four or Ave "different kinds of moths and prob-
ably clethra by as many more ; and one of the kinds which
visit privet, Oenotliera, and Azalea viscoi^a^ I have seen
visiting the scentless purple larkspur at dusk when I could
scarcely distinguish the flower from the leaves ! and what is
of farther interest was the fact that a few white flowered
varieties in the clump were passed by ! Datura Tatula is
said to bloom towards evening (it is of a pale violet-[)urple),
and if so, there is no doubt but that moths would Hnd it
with its strong odor as readily as larkspur !

Apoc]/num androscmiifolium, a day bloomer, is a bell-
shaped flower one-third of an inch long, of the palest pink,
with fifteen deep rose-colored lines running straight to the
centre of the flower. Two different kinds of moths were
noticed at dusk sucking its nectar, not counting diurnal
insects.

Honey guides are useful characters in distinguishing-
species. Of the two New England yellow-flowered species
of Oxah's, viz., O. striata, L., and 0. cormculata var. Dil-
lenii, Trelease, the latter can be distinguished at a glance
l)y the honey guides. They are very diminutive, consisting
of four or five straight pencil lines near the base of each
petal, forming a small circle around the centre of the flower.

When an animal seeks a lirook or spring for water, he is
apt to take the shortest practicable course. If he goes and

74

returns several times, he is likely to make a path. Figura-
tively speaking, the special markings on the petals of
flowers are the foot-prints of the bees and the butterflies.
When they follow the same route for nectar, they leave a
trail ; where the butterflies walk around the stamens of a
pink they leave the impressions of their tiny feet in the
shape of a circle ; when a bee rubs his back against the
petals of a poppy he makes a similar shaped spot on each ;
and when with his tongue he laps the nectar on the petals
of the tulip-tree, he paints a golden band ; when he falls to
the bottom of the cup-shaped tulip, he in his scrambling
describes a disc of blue or yellow.

Bees and butterflies work unconsciously as Nature's agents
in flower decoration ; some of the markings prove to be of
utility in guiding the insect descendants of those who made
them ; some, in regular flowers Avhere the lines from every
side [)oint to the centre, arc of no especial value ; others
neither direct nor have any connection with the nectaries,
but all, l)eing footprints of nectar loving insects, may pos-
sibly indicate to the more intelligent of their kind that the
flowers are especially nectariferous.

VARIEGATION OF LEAVES, ETC.

Special markings are not peculiar to flowers ; the varie-
gation extends to leaves and steins. Notice the pale cres-
cent on Clover leaf ; the purple zone on Pelargonium zonaJe;
the mottled leaves of Erythronium and Chimaphila ; the
brown-purple spot on Enphorhki niaculata, and the more
brilliant markings on leaves of tropical plants, referred to
elsewhere. As the upper and lower surfaces of petals are
sometimes of different colors, so leaves are found with one
side differing from the other ; some species of Oxalis, Tra-
descantia, Begonia, Nymphani, Cineraria, etc., are reddish-
purple beneath and green above ; certain species of Quercus,
Populus and Vitis are white beneath. On trunks and stems
of trees and shrubs the green or brownish basal color is

75

often prettily variegated with nunierou.s white dots a.s in
birch, willow, alder, sassafras, benzoin, etc. ; pale gray
blotches ai)pear on some mai)les, and larger white patches
on Platanus, while the whole trunk of canoe and white
l>irch with age becomes a staring, chalky white.

SEEDS.

The colors of seeds take a range nearly e(iualling that of
flowers, every hue being found from the white of pumpkin
to l)lack of watermelon. The garden bean, perha{)s, pre-
sents the greatest variety of colors and variegation in its
different kinds ; it is quite astonishing, as these seeds do
not receive the direct rays of the sun, l)eing enclosed in a
pod, and yet they are often })rettily marked. Indian corn
is red, yellow or blue.

The order of development of seed colors as in flowers, is
not uniform. Notice a few examples from small seeds :

SteUaria media : the transparent ovule flrst becomes
white, then orange, brown, and black in succession.

Sonchui< oleraceous, goes from white through lemon-yellow
to brown.

Oenothera biennis, from white to brown ; no yellow.

Bcqytisia tincloria, from green, through pur[)lisli-brovvn to
black-purple.

Asparagus, from a translucent green directly to jet-l)lack.

The colors of different berries beginning usuall}^ with
green, show many beautiful transitions before reaching their
flnal hue. The gradations of Asparagus berry for instance,
are green, olive green, Indian red to scarlet. Others
change from green, through red to blue, i)urple, or black;
green directly to vvhite ; green through yellow to scarlet ;
and white through green to scarlet (Gaultheria), etc.

CHAPTER XII.

ARE THE COLORS OF FLOWERS THE RESULT OF INSECT

SELECTION ?

Prof. Hermann Mliller, in his book on "The Fertilization
of Flowers," ascribes their color to the selective agency of
insects, which having preferences for certain hues, have
visited such more frequently, and consequently have pro-
moted their increase by cross-fertilization. We quote from
him as follows :

"In Rosacem, whose honey lies concealed and which are
fertilized by a motley crowd of short-lippod insects, the
flowers are for the most part greenish-yellow, yellow, or
white in color. In Gomarmn palustre, Potentilla atrosan-
guinea and Sanguisorba officinalis, they are dark red or
pur[)le, probably owing to the influence of carrion-feeding
flies, etc. Rosaceoe visited for pollen have white flowers
when the chief visitors are small, short-lipped insects."

"The uncommon color of Scro^iJndaria must be referred
to the peculiar taste of its visitors, the was})s. It has round
brownish flowers with widely open mouth."

"Bright red colors of pinks seem to have been produced
by the similar tastes of butterflies."

"Flowers fitted for short-lipped insects are usually white
or yellow."

"Insects with longer tongues and acuter color sense
gradually caused the production of flowers with more varied
colors."

" Globular ia is blue. This is the only instaiu-e in the
German and Swiss flora of a l^lue color h^xwg produced hy
the selective agency of Lepidopienr (butterflies).

"Hundreds of ht)ney bees visit both Melilolns o^cinalis
(yellow), and 31. vulgaris''' (white).

77

''Linarla vulgaris, flower yellow, excludes shoi't-lii)i)ed
l)ee8 from the honey, find flies and beetles are prevented
from entering the flower by the tumid lower lip whieh
completely closes the tube ; by these characters the flower
becomes exclusively adapted for the most diligent of fertil-
izing agents, the long proboscisid bees."

"Antirrhinum majus, })urple, our largest l)umble bees
can enter bodily. It is fertilized chiefly l)y bumble bees."

" DigilaUs purjmrea, huml)le bees are the only fertilizers" ;
flower purple.

"Oenothera biennis, honey Mcccssible to long-tongued
bees, and also adapted to nocturnal Lepidoptei'a," flower
yellow.

" Galeobdolon luteum, yellow, exclusively by bees," seven
species.

Clematis recta, white, seven species bees out of nineteen
species insects.

Stellaria media, white, six s[)ecies bees and Ave diptera.

Chelidonium majus, yellow, seven species bees and six
diptera.

Ranancidus acris, yellow, twenty sj)ecies bees out of
sixty-two total species.

Cnicus arvensis, i)urple, thirty-two species bees out of
eighty-eight total species

Taraxacum offi,cinale, d.-indelion, yellow, flfty-eight spe-
cies bees out of ninety-three total species.

Tf one analyzes the foregoing ({notations, it will be seen
that Midler says distinctly that the coloi-s of certain Howers
are due to the peculiar tastes of their insect visitors, who
by their selective agency have produced them.

That some plants have changed the colors of their flowers,
and that others are in ])rocess of change, we have ah'eady
shown ; our object now is to ascertain, if possible, whether
bees have a preference for certain colors and are able at
their own sweet will to change one hue for another, or
whether the plant itself, for its own advantage, owing pos-

78

sihly to changed conditions, initiates, carries on, and com-
pletes a change in which the insects are only valuable
assistants.

As a means of solving the question, we have taken note
of the principal visitors of a hundred different kinds of
flowers, more or less.

We find that it is a difiicult question to determine posi-
tively, owing to different conditions in different localities ;
the species vary, the seasons vary, and the insects vary ;
competition varies, and the daily weather varies ; but we
are able to say what flowers certain insects hav'e visited and
in which they appear to delight. Our attention has been
directed principally to the larger kinds : bumble bees,
honey bees, wasps, hornets, butterflies, etc., while not
entirely neglecting the minor insects. We will anticipate
our conclusions by saying that taking a general survey of
the subject we cannot And any marked preference of any
kind for any one color.

In the Fall of one season, when golden rod and Joe-Pye
weed (Enpatorium) were at their height, these flowers
bloomed a))undantly along the side of a road leading through
swampy woods. Bumble bees were present in great num-
bers, visiting both of these flowers, one kind yellow, the
other i)urple. An actual enumeration was made, and the
number of bees on one kind was substantially the same as
on the other. The following year, in a different locality,
there were bumble bees found on Joe-Pye, but scarcely one
on goldenrod. The only way I could account for the differ-
ence was by supposing that the bees of the woods, l)eing
remote from flelds, were restricted to woodland flowers,
while in open flelds they would have a greater choice of
kinds.

Symphoricarjjos vulgaris was discovered in Eastern Mas-
sachusetts. The Floras do not give the plant as growing
in New England, so that likely few, if any others, could be
found within two hundred miles. The flowers were monop-

79

olized by hornets to the cxckision of every other insect.
There were more hornets assembled than I had seen together
at one time. The tlowers were small, inconspicuous, bell-
shaped and perfectly green. Would it be safe to say from
this incident, that hornets prefer a green color and a bell-
shaped flower? If they do prefer it, and our Symphori-
carpos is not to be found within two hundred miles, what
are the thousands of hornets to do at this season of the
year? it blooms in August. Evidently they must be con-
tent with other colors, green flowers being scarce, or conflne
themselves to killing insects.

I have seen hornets on garden leeks having a similar
shallow, liell flower with colors from green to purple; also
on the small green bell flowers of Gai/lmsacia frondosa,
but as a matter of fact I have seen more hornets on other
colors than on green. Referring to my memoranda, they
were found on Symphoricarpos, Eupatorium, Clematis,
Rhus, Helianthus, Linden, Rudbeckia, Clethra, Monarda,
Asclepias, Solidago, Ribes nigra, Spira?a salicifolia, Gay-
lussacia frondosa, etc.

On the other hand, I have examined numerous other
green flowers without ever seeing a hornet.

I infer, therefore, that the hornets were especially pleased
with Symphoricarpos, on account of its abundant nectar and
the ease and convenience with which it could be ol)tained.

I, of course, do not pretend to an exhaustive research in
regard to hornets. Doubtless if my observations were
continued, other green flowers would be found to be fre-
quented by hornets, and at the same time the list of colored
flowers would be extended.

We do not find in studying the habits of the larger winged
insects that any particular color exercises such a fascination
as to attract especially to the neglect of others of a different
hue. Mliller maintains that white and yellow flowers arc
frequented by inferior fly-like insects, but that the nobler
honey bees and bumble bees have a preference for purple or

80

blue. Of course any bright color in contrast with the
green foliage Avould arrest attention, but examples of green
tlovvers being freelj' visited by swarms of insects are too
numerous to warrant the assertion that color is essential,
and especially a particular color for particular species.
There is not the slightest doubt, however, that bees do
distinguish between colors and that they i)rofit by the dis-
tinction, as Avhcn they select, not from taste, but for their
advantage, a white rose with yellow stamens rather than a
white rose with brown stamens, or a white Weigelea rosea
in ])reference to a pink one.

Moreover they have no choice in the matter, for if they
wish the freshest and best supply of nectar they must take
the lirst color in these changeable flowers, in one case
yellow, in the other white, etc. ; in Lathyrus maritimus and
numerous others they are obliged to take pink which is
followed by blue, and in Ipomu'a, morning-glory, and
Brunella, violet followed by pink !

Again, while bees as a rule visit one kind of flower at a
time and then of course those of the same color, if the
kind of flower is plentiful, in case it is not they go from
one kind to another indiscriminately, as we have seen
bumble bees fly from Brunella to Baptisia, from Impatiens
to Arctium, Nabalus to Clethra, and Brunella to red Clover.
Later in the day they take another sort of flower which they
follow, and as the season advances the same bees, or same
kind of bees, follow in succession five-sixths of the different
flowers as they appear, of whatever color.

We will notice some of the flowers visited by bees and
some of their preferences for certain kinds of flowers, not
preferences for mere color.

PURPLE AND BLUE.

Brunella and Pontederia are visited mainly by bumble
bees, some honey bees and butterflies ; these also visit
Viola, Iris, Lupinus, Delphinium, Aconitum, Mentha,

81

Asclepias, Cnicus, Eupatorium, Echiiim, and other Borra-
genaceiv, all purple or blue, but the same bumble bees visit
the red ch)ver (Trifolium), which is of a distinct |)ink-
pur})le, just as freely, while honey bees, on account of their
smaller size, are obliged to take the white T. repens, or
rose-tinted T. hybridum, which have shorter tubes. Honey
bees are constant visitors of Myosotis, blue ; l)umble bees
visit less frequently ; on the other hand buml)le l)ees visit
pansy, a larger Hower, purple or yellow, more freely than
honey bees.

Wistaria, blue, has j)ea-shaped Howers which exclude
many kinds of insects; bumble bees, but more especially
Xi/Iocopa, an insect quite as large as the largest bumble
bee, with a smooth l)lack abdomen, seeks them liefore they
are fully expanded and pierces the short tubular calyx from
the outside for the nectar, and when fully blown they also
visit the interior. Honey bees suck the nectar from the
minute holes made by the former. Bumble bees often
employ this device to get at the honey in other species of
various colors. I have seen them pierce the tubes of Del-
phinium, Rhododendron, Azalea viscosa, Weigelia rosea,
various varieties of Petunia, Aquilegia, Vaccinium eorym-
bosum, Impatiens fulva. Phlox, Foxglove, Leucothce, and
Linaria vulgaris, even in bud. All hues are represented in
these flowers : red, blue, yellow, white and green I These
flowers do not appear to be perfectly adapted to bees, and
they have considerable difl[iculty in reaching the nectar, yet
they do not turn their backs on them to seek purple or blue
flowers, although it is said, " blue is the special hue affected
by bees." Honey bees do not attempt to enter these flowers
except Weigelia rosea, they know what the bumble bees
have done and without any honey guides they go straight
to the minute holes on the outside of the nectary !

While bees are conspicuous on the purple flowers above
mentioned, many of a similar color are not visited by bees
at all and scarcely by any insects, as the large purple O/e-

82

malts, Vinca, Ageratiim, Sisi/rinchkim, the smaller Lobelias,
Solanum Dulcamara, Agapanlhus lonbeUafus, and most
probably other l)lue lilies, Hydrangea hortensis, Tradescantia
Virginica, etc.

WHITE.

Among white flowers our native and garden varieties of
Clematis are visited l\y numerous honey bees which lick the
juices from the filaments and anthers, as the flowers have no
nectaries. Clethra, white, is visited by many large size
insects attracted by its fragrance as well as nectar, such as
bumble bees, honey bees, wasps, hornets, ichneumons, drone
flies, l)utterflies (yellow, black, or brown-eyed, etc.), hum-
ming-bird moths and other moths, and the ambush bug, etc.
Of this promiscuous crowd the l)ees are always very umch
in evidence. Notwithstanding the attractions of this sweet
flower, the purple Asclepias, the Canada Thistle, the crim-
son-purple Burdock, and the pale green flowered Rhus
copalh'na attract just as well both kinds of bees, and about
the same mixed assembly.

Ligustrum, the common privet, also white, with small
tul)ular flowers, is visited while the day lasts and until the
last corolla falls to the ground, by both kinds of bees and
numerous butterflies and drone flies. Melilotiis alba, a
small, sweet-scented flower of the Leguminosm, is much
liked by bumble bees. One will rarely miss finding numer-
ous heavy-bodied bees clinging to the slender pea-shaped
flowers ; the same may be said as to Hydrangea paniculata,
Cephalanthiis, Deutzia, etc. ; on the other hand no bees ever
visit Sambucus Canadensis, PhiJadelp)hus inodorus, Solanum
tuberosum, or Exochorda, all very showy white flowers,
having no nectar, but a plenty of pollen. Our white flowered
cherry trees, pear, quince, and apple trees, the latter more
or less tinged with crimson, are visited by throngs of honey
bees and many bumble bees, which aid greatly in fertilizing
the flowers, also Prenanthes, CepJiaJanthus, Hydrangea pa-

83

)iici(Ja(a, Aster iinibeUaius, Nym^^hct^a, Ihihsia, ^Escirhis,
white lilac, white foxglove, white rose, niigiioiiettc, etc.,
all white flowers, are frecjiiented some by honey bees, others
by bumble bees.

Bees do not care for the small, Hat flowers of spiriea,
strawberry, viburnum, cornus, sambucus, achillea, May-
weed, etc., all of which are white. They also discriminate
between flowers of the same family and same color; Rubus
strigosus and R. occidentalis, rasp])erries, have diminutive
white flowers much freciucnted hy all kinds of bees, while
the larger and showy R. villosus, high blackberry, has but
a limited number, certainly many less.

It appears to one, at first, singular that bumble bees which
are crazy for the white, fragrant flowers of Rosa rugosa,
var. alba, will never alight on the white, sweet-scented and
beautiful flower of the water lily, Nym})hiva. I sus})ect
that they do not like the water, and are afraid of wetting
their feet; besides, there does not appear to be any nectar,
and the limited amount of pollen is better adapted to small
Hies, yet honey bees enjoy these flowers.

Bees are so fond of white privet that I have often seen
them (bumble bees) remain so long at the close of day that
they went to sleej) on the flowers and remained there all
night !

RED OK KEU-PURPLE.

Bumble bees are fond of red clover, which, as has been
noticed, is of a very different color from Brunella and Viola.
All of these flowers are more limited in kinds of visitors, on
account of their peculiar shapes. Pink-purple foxglove, as
well as all other tints, is visited by both kinds of bees ; the
bumble bees readily entering the tube, not regarding the
very slender hairs almost invisible on the lower lip, which
cause the more sensitive honey bees to turn and enter, if
they can, by the upper side. No more bees visit the flaming-
red flowers of Pyrus Japonica than the pure white cherry

84

l)lossoms, both of the same family and similarly constructed ;
no more a red rose than a white one.

Honey bees revel on hawthorn, while bumble bees rarely
ap{)roach it. The scarlet or red tulip is mainly visited by
honey bees.

Rhexia Virginica is visited by both bumble and honey
bees. Bumble bees can usually be found on RJiododendron,
^Esodiis, Rohinia, AquiJeyia, Weiyelia rosea, Zinneu, Gladi-
olus, Hibiscus, scarlet poppy, foxglove, balsam, rose, scarlet
salvia, dahlia, malva, Apoci/mim, Polygcda, Gerordia and
Elodes camimnidata, etc. These are mostly large flowers
and tubular.

Si)ircea salicifolia, flesh color, is frequented by bees and
many other insects, while S. tomentosa, rose color and more
showy, has almost none. No bees have been seen on Kalmia
latifolia and K. augustifolia. It is said that bees do not
frequent oleander (Nerium Oleander).

Bajjtisia tinctoria, Cytisus scoparius, and probably all
the larger yellow papilionaceous flowers are visited l)y
bumble bees quite as exclusively as Wistaria, Brunella, or
red Clover.

Goldenrod is much liked by honey bees and other small
bees vvhich can always be found on it, while l)uml)le bees
not so uniforndy ; wasps and hornets are also often seen
on these flowers.

Taraxacum and Rauuncidus hidbohus have many honey
bee visitors, but a limited number of bumble bees. E. acris,
with only a minute nectariferous scale, has no l)ee visitors.

IleJianthns annus attracts more bumble bees than honey
bees ! also hornets and several other kinds of bees.

Asdepias luberosa, yellow or orange, many ])umble bees
and honey ))ees, various kinds of butterflies, common flies,
etc.

Puin})kin, squash, melon, and cucumber, all yellow, are

85

fertilized almost exclusively by hianble tmd honey bees.
Five or six honey bees may sometimes be found in a single
|)um[)kin flower, scrambling over each other.

Oenolhera biennis by bumble bees and other bees, a i)ink-
and-white moth and another of a dusky color.

Irn2)atiens fulva is visited by bumble bees who I)oth enter
at the orifice and afterwards bite holes in the shar[)-turned,
fish-hook spur from the outside, to get every droj) ! also
by hunmiing birds.

IL/l^ericiim and Hypoxia rarely have insect visitors ; a few
honey bees oidy have been seen on Hypoxis.

Ribes aureum, yellow, with a long calyx-tul)e, is visited
by bumble bees.

Linaria vulgaris, yellow, is visited exclusively by bumble
bees and another small bee.

Silpliium asjierrinifun and lleliopsis scabra are two i)lants
of about the same height, with yellow fioweis, similar in
appearance. The former is visited by many bumble bees,
while Heli()j)sis has none. The florets of Silphium were
longer and looser ; those of Heliopsis shorter and more
compact ; these were the only visible differences. Heliopsis
was not entirely forsaken by insects, however, for it had
several kinds but smaller in size.

Sassafras has small greenish-yellow flowers visited by
flies and ants. Bees do not visit ihem. If they alight
casually and taste the nectar, they innnediately leave in
disgust.

Tilia, on the other hand, of a similar color but even
paler, has throngs of bees and other insects which all to-
gether make a distinct hunmiing sound I

Oakesia sessilifolia is of a dull cream color, flowers droop-
ing, under overspreading leaves. The flower is often
scarcely visible, and yet bumble bees will fly close to the
ground, passing violets, potentilla, strawberry and other
flowers, to find this bonne bouche.

Bees are also connnonly seen on Helianthus giganteus,

86

Marygold, Helenium autumnale, Actinomeris squarrosa,
Laburnum, Berberis, Aesculus glabra, A. ttava, Crocus,
TropaH)lum, Caniia, Gerardia, Cytisus, Linaria vulgaris,
Solidago, Impatiens fulva, Cnicus horridulus. Ribes aurea,
yellow, with long calyx tube, is visited by bumble bees, but
they never visit the myriads of Potentilla Canadensis, same
color, and rarely the more showy Forsythia. They also
discriminate between dandelion and buttercup, and seek
out Hypoxis from among Potentilla, Oenothera pumila and
buttercup !

By the 1st to 15th of June many flowering shrubs, fruit
trees, etc., are past blooming, and bright colored flowers
are for a time rather scarce ; then, many dull, green colored
flowers appear, and one unfamiliar with the fact would be
surprised to know to what extent these inconspicuous flowers
are visited by bees. In several instances while walking-
through thickets, not noticing anything but green leaves,
my attention was attracted by the humming of bees. They
were found to be in great numliers on the perfectly green
pistillate flowers of Nyssa sylvatica in one instance, and on
Gaylussacia frondosa in another. These bees had left /r/.s,
Sisyr'mchium, Cnicus horridulus, Kalmia, blackberry, but-
tercups, etc., and for the sake of a green flower? Not at
all. The nectar of the flowers just coming into bloom was
undoubtedly the chief attraction. For several years I have
observed the fondness of honey bees for the spicy-scented
Ptelia trifoliata, and have awaited the time of its flowering
to see the thousands of clustered green flowers about to
blossom, visited by many honey bees eagerly watching their
opening, and that as long as a blossom remained they con-
tinued their visits. Amjjelopsis Veitchei, green with small
flowers, often entirely covered by the large-leaved foliage,
petals and stamens soon falling, yet they are much visited
l)y honey bees !

87

In the first half of June the following green flowers bloom :
Petelia, StniJax, GayJuxsacla frondosa, Vitw, Liriodendron,
Gymiiodadvs Canadensis, Ehamnus cathartica, Nt/ssa si/l-
vatica and Rhus Toxicodendron, and all hut the last are
visited by honey bees. They are also very fond of lilnis
glabra, B. typMna, Rihes nigra, English Ivy, Ampelopsis
qninquefolia. Asparagus, Andromeda, Salix, Norway Maj)^,
and, notwithstanding its disagreeable odor, Ailanthus glan-
duJosus. The latter was visited by both honey and bumble
bees, but bumble bees were rarely or never seen on the
other green flowers excepting English Ivy, Gaylussacia,
and Andromeda.

A recent writer observes that, "Thistles are purple to
please the bees." If so, what insect w^as intended to be
pleased with the yellow thistle? Certainly bumble bees
appear to be greatly pleased with them and are by far the
most numerous visitors.

One finds a marked difference in the habits of honey and
hum))le bees. The former take a much wider field of opera-
tion. They visit readily nearly every green flower exce})t
the veriest weeds, beginning with the willow in the Spring,
and continuing until the English Ivy blooms in the Fall.

Bumble bees, on the contrary, visit but a few of tiiis
color, not even Liriodendron, preferring the brighter hues
and tubular form, yet some of bright colors and favorites of
honey bees are not frequented by bumble bees.

If any kind of bee has a preference for blue it is the
bumble bee, but those flowers of a blue or purv)le color that
he visits the most frequently are always visited also by
many kinds of butterflies, which proves that those flowers
are rich in nectar. Also when one sees the white cabbage
butterfly, Pieris, and the green CoJias, supposed to be es-
pecially fond of Brassica, delighting in Pontederia, Helio-
trope, Brunella and other l)lue flowers of same family, it is
quite evident that butterflies also have no preferences for
color. I have seen the same kind of dusky moth visiting

88

in the evening, Oenothera, yellow ; Azalea viscosa and
privet, both white ; Apocynum androsrem(foh'u7n, pale pink ;
and a garden larkspur, var. candelabrum, blue-purple !
One is not likely to know what its preferences Averc for
color, but it certainly appeared to l)e enjoying itself as
much on one as on the other.

It is the same story for the moths as for the bees and
butterflies, wasps, hornets and ichneumons, they all seek
honey wherever it can be o])tained, color or no color ; and
quite a percentage of the smaller kinds are content simply
to sip the honey-dew from the green leaves, never visiting
a flower, as is easily proved by walking through bushes in a
midsummer evening where there are no flowers, and seeing
the millers fly up before you on every hand, to alight again
on other leaves !

In a comparison of Alpine Gentians, Muller suggests the
following theory of their evolution :

G. lutea is of a lower grade than the others and nearer
the primitive form. From one branch of the original a
campanulate form was evolved in its relation to honey l)ees.
Finally the corolla became so narrow that Lepid()})tera as
well as humble l)ees were obliged to perform cross-fertiliza-
tion. "The j)rimitive color retained in G. lutea, was grad-
ually exchanged for blue by the influence of the huml)le
bees."

When one notices bumble bees so hardly pressed in the
" struggle for existence," that they bite the sticky tubes of
white Azalea viscosa and the yellow tubes of Impatiens and
numerous other sorts of tubes which it is impracticable for
them to enter at the mouth, one can hardly believe that
these same bees are all this time diligently developing blue
flowers, and that wasps arc developing yellow flowers when
they seek blue larkspur from which to pilfer the honey in
the same way, each one abandoning his supposed favorite
color to obtain honey with difficulty from another which he
dislikes ! But in addition to the evidence before our eyes

89

in our gardens and fields of no preference for a particular
color, aside from favoring forms, Mliller himself, besides
the instances above quoted of yellow flowers being exclu-
sively cross-fertilized by bees, states as follows: "The
su]})hur-yellow color of Semperviviim Wulfenii seems not to
stand on the same rank as the yellow color of some Sedums,
but rather to have been developed from a puri)lo color l)y
the selective influence of huml)le bees" I

"In several species of Lonicera fertilized l)y bees the
colors are bright red " !

"In L. caerulea, adapted for humble bees, they are yel-
lowish-white " !

Mliller clearly is impartial in his examples when he gives
his opinion that purple w-as developed from yellow in one
instance by bees, and in two others, bees developed yellow
from purple !

Lubbock experimented with different colored slips of
paper pasted on glass, upon Avhicli he put a drop of honey,
and put them on a lawn for the bees, transposing the colors
from time to time, with the result that the bees showed a
decided preference for blue.

Mliller on the contrary gives a table of actual visits of
bees to different colored flowers, showing that those of a
yellowish-white received the most visits, and blue the least
number I

We have seen that bees discriminate between flowers of
the same color in the same genus, and between flowers of
the same color in different genera, also between flowers of
different colors. They actually prefer some small green
flowers to some showy blue flowers, and vice versa ; and it
is clear that their preferences are for the flower and what it
affords and not for mere color. When one sees honey bees
leaving beds of red begonia and pelargonium or blue myo-
sotis for white clover close cropped by a lawn mowing
machine, then forsaking clover for green Ptelia when that
comes into bloom, he is quite ready to declare that azure-

90

loving bees only love an azure color when it is for their
interest. I have even noticed bees to leave flowers for a
time and be content with only leaves !

There was a humming sound of bees in a purple-leaved
beech tree, and searching for the cause there was found to
be on the under side of many leaves a white, woolly sub-
stance which concealed numerous green aphides, and that
they were living on pellucid little drops of a gumm}^ nature
which also attracted the bees ! Similar sweet secretions are
found on the common Fagus ferruginea, elm, and in less
degree on red ma})le. It is a wonderful faculty that bees
have which will enable them to find such minute secretions
on leaves, or on inconspicuous green flowers without per-
ceptible odor. It is scarcely conceivable when one is
acquainted with the hal)its of l)ees that will so diligently
and indefatigably search for nectar, that they would hesi-
tate on any particular color, because, forsooth, it was not
the right tint I

Bumble bees are a sturdy race of insects, made to crowd,
push, probe, and burrow ; therefore they prefer a tubular
or bell-shaped flower that they can enter, or a personate or
papilionaceous flower that they can force, or a tubular
flower not too long that they can probe. Naturally they
do not care for small flat flowers or diminutive florets of
composites, such as ox-eye daisy, erigeron, achillea, etc.

The minor insects, as might be expected, fly indiscrim-
inately from color to color without preference, arid the
same is the case with l)utterflies. I have counted more
than twenty difterent kinds of flowers which were visited
by the same species of butterfly ! And about the same
number by wasps. There is probably not a single kind of
flower but that is visited by half a dozen kinds of insects
and probably the greater part are visited by many times
that number.

In the case of the tulip we showed how the same bee
produced three different colors at the same time by stimu-

91

luting the petals to increased action. In like manner not
only bees, but every other insect is doing his part to stimu-
late the flowers to produce more and brighter color. They
do not select, they simply stimulate, and the flowers or
plant does the rest. Regarding the prevalent o{)inion that
bees have a preference for blue colors, the writer has
experienced more difiiculty in flnding bees on blue violets,
sisyrinchium, iris, etc., than on dandelion and buttercup.
I have never seen a bumble bee on heliotro})e and they do
not visit larkspur with much zest, almost always biting-
holes on the outside. I have seen more ichneumons on
Asclejnas Cornuti than bumble l)ees ; seven large, black,
waspy ichneumons were seen at one time on a single umbel.
As between Antirrhinum majus, purple, and Linaria
vulgaris, yellow, being similarly shaped, there is no prefer-
ence whatever. Both arc fertilized by the same bees, viz.,
bumble bees, and another undetermined species rather
smaller than a honey bee, yellow or brownish on under
side of abdomen. While the former could only enter half
way into Antirrhinum, the latter pushed entirely into the
flower which closed after him. He was able to turn around
and emerge when he chose

One can count more confldently upon flnding bumble bees
on l)utter-and-eggs, sil[)hiun:, baptisia, Helianthus animus,
etc., than on any of the above or Agapanthus, Funkia,
Centaurea Cyanus, etc. Yet certain families of blue are
nmch frequented by bees, as the Borraginaccic, Labiata',
Bontederia, etc.

When a number of favorable conditions combine in a
l)articular family or species, then bumble bees appear to
have a preference for the flower. These conditions exist in
Anchusa, viz., size and shape : tubular flower one-third inch
long; throat slightly closed by feathery processes, (the
l)ees enjoy a Httle obstructicm), a continuous supply of high
grade nectar, and a bright blue flower ! The bees visit
these flowers over and over again without intermission from

92

morn till dewy eve. Very similar conditions exist in all
the Borrauinacetv, in Brunella and Pontederia, but when
like conditions meet in white privet, or clethra, the bees
are just as numerous ; also in a different style of tubular
rtower as crimson or white foxglove. This last, of a totally
different color from Anchusa, never lacked for buml^le bees
while a flower remained on the stem

Honey bees have absolutely no regard for special colors ;
they go readily everywhere that nectar is accessible, natu-
rally taking shorter tubes to probe than the bumble bees.
A green flower is just as acceptable to them as a blue one,
provided it is nectariferous, their business being to collect
honey, and if green flowers can furnish it, or even green
leaves, they will be on hand.

Moths visit at twilight yellow, white, green, red and
pur})le flowers equally ! The same humming bird was seen
to visit the following flowers in the order stated : Gaillardia,
Digitalis, Campanula, Heliotrope, Lonicera, Gladiolus ; all
colors represented. They also visit Tecoma, Tropa-olum,
Arctium, Impatiens, Tiger Lily, etc.

There are of course certain kinds of beetles and bugs that
feed upon the foliage or flowers of a [)articular family as
upon Asclepias, Solanum, etc., but these care not at all
for color.

The only difference I And regarding white flowers may
be explained by their size : Mayweed, erigeron, achillea,
slum, daucus, etc., have the very smallest white flowers,
and very naturally would be visited by flies, the most di-
minutive bees, and the smallest butterflies, as Clivi/sojjJianus
and Fhyciodes, and this is exactly the fact in regard to
these flowers ; l)ut when the same colored flowers get to
the size of clethra, privet, cephalanthus, garden chei-ry,
blackberry, white pond lily, etc., which are no more spe-
cialized than the former, the bees are very much in evidence
and also the larger butterflies.

There are a dozen kinds of bum])le l)ees, Bombiis, in

93

New England, and I htive noticed considerable difference
in the various species as to tlie llowers they visit. There
are also over one hundred species that are classed as bees,
large and small, in the same territory, and taking together
all the tlovver-visiting kinds of insects, their number would
be in the thousands !

In a i)ond I found a large quantity of Pontederia, pick-
erel-weed, while the border was lined with Cephalanthus.
The pickerel-weed was visited by bumble bees and several
pretty maroon or reddish-brown banded humming-bird
moths and several kinds of butterflies, but by no honey
bees ! Cephalanthus was visited by many honey bees and
a number of bumble bees, also butterflies. The white
flower of Cephalanthus, just coming into bloom, was pre-
ferred by the honey bees to the 1)1 ue of Pontederia !

At about the same time it was observed that there were
scarcely any honey bees and but few bumble bees in a
garden of twenty or more choice varieties of flowers, several
of which were pur})le or violet. The apparent explanation
was, that at a short distance from the garden there was a
lawn containing a number of bushes of the sweet scented
l)rivet, which apparently had attracted all the bees o^the
neighborhood.

A little later in the season every honey bee deserted
heliotrope and mignonette, which earlier had been especially
attractive to them, for Amj)elo2}sis Veltchel, a perfectly
green and insigniflcant flower, almost concealed by broad
leaves ! They even appeared to |)refer this flower when
petals and stamens had dro})ped off. Bumble l)ees did not
api)ear to have the instinct to Hnd it. These examples
from the pond and garden illustrate the competition of
flowers and the changing tastes of insects, and also prove
that there are greater attractions than color.

In relation to the color of a flower being adapted to its
environment, I have remarked that scarlet tulips expand
their cup-shaped corollas sooner after cold or cloudy weather

94

than white or yellow of the same age in the same bed ; and
that the latter close earliest at the approach of a cold storm.
This appears to be due to the difference of temperature in
the flower; the white requiring most heat, the yellow less,
and the red least. The pink Cornus florida blooms a little
earlier than the white.

Newbigin states that "the red parts of plants absorb 1.82
per cent, more heat than green parts."

Another observation may be worth noting in this connec-
tion, viz. : I have remarked that in the Southern States
there is a preponderance of yellow flowered si)ecies in sev-
eral genera which are common to both the North and the
South. Following is a partial list :

Chapman's

Botany.

Gray

's Manual.

Polygala,

5 yello

\v species,

17 purple.

2 yello

vv, 12 purple.

C'rotalaiia,

3 "

none "

1 "

none "

Aeschynonieu:

I 2 "•

u u

1 "

u u

Baptisia,

9 "

" "

2 "

" u

Aqrinionia,

3 "

" '•

2 "

u u

VValdsteima,

2 "

u u

1 "

ct u

Rhexia,

1 "

u u

0 "

3

Nymphaea,

1 "

u u

0 "

none "

Aqyilegia,

2 "

u u

0 "

1 scarlet, 1 blue,

Sophova,

1 "

U (I

0 "

noue purple.

Astragalus,

2 "

" u

1 "

" i'

Rhynchosia,

6 "

and 1 var

3 "

" •'

Thermopsis,

3 "

u u

2 "

" "

Cassia,

7 '■

aud 1 var

4 "

U ti

The species descril^ed by Gray as greenish, cream color.
Chapman calls pale yellow.

California furnishes yellow Iris, yellow Lupinus, Esch-
scholtzia, Calochortus, etc. DelpJiinimn cardinale and D.
nudkaule have orange or scarlet sepals or })artly yellow
petals, while Northern species are })urple. It would not be
at all improbable that some yellow-flowered species originat-
ing in Southern climates should on spreading farther
North become flushed with red or purple.

95

AN INSECT WITHOUT TASTE FOR COLOR, AND WHICH
PREFERS BLOOD TO SWEETS.

Phymata Wolffii, the Ambush Bug, is a common insect
on the flowers of the tield and garden. He is a deadly
enemy to honey and bumble bees, large flies and l)utterflies.
The appearance of a spider, so different from a winged
insect, would seem to be enough to frighten away any of
the latter and prevent their capture unless taken unawares
in a web, but this demure, winged bug, only two-fifths of
an inch long, dressed in quiet yellowish or greenish colors
and marked on its back with a brown-colored cross, gives
no warning of its sanguinary nature. It is quite innocent
looking, :ind appears to have as good a right to visit the
flowers as bees or butterflies. It sometimes hides under
the petals, especially when they are small and clustered,
])ut quite as frequently takes up a conspicuous position on
the flower, where it remains perfectly motionless until the
legs of the bee or other insect are in a favorable position,
when they are seized by the diminutive but powerful clul)-
shaped claw of the bug, stabbed in a vital part l)y the
dagger- like proboscis, and their life blood sucked away.
In one instance the bee's tongue was seized and by that he
was held. It is not pleasant to see a noble insect like a
bee, or a beautiful butterfly, or even a drone fly, sacrificed
by this worthless thug with the sign of a cross on his back.
it is needless to sa}' that the nuirderer is not fastidious as
to the color of the flower he visits: yellow, white, red or
blue are all the same to him, provided there is a good
supply of victims !

CRUELTY TO INSECTS.

The structure of the flo\ver and the manner of pollination
in Asclepias is as complicated as in orchids. The five deep
wedge-shaped slits in the crown are veritabl(> death-traps to
some insects. While spiders have I)cen seen to walk over

&6

them with impunity, liumble bees invariiibly get their legs
entangled, but by a vigorous jerk they are able to pull
them out, and often with the pollen masses attached.
Wasps are more awkward, and twist themselves round and
round, but finally escape ; while flies and miller-moths often
become hopeless prisoners and an easy prey to prowling
s[)iders. Sometimes they succeed in their struggles for
liberty by wrenching themselves from their pinioned limbs,
which are left standing on the flower as monuments of a
freed but mutilated body. A dozen flies have been counted
on a single umbel, all securely entrapped by these cruel
pollen pockets. Ten times as many honey bees as bumble
bees visit this flower, at least in the vicinity of towns, and
they appear to be very efficient fertilizers too, yet I have
found several dead, held fast by a leg, when there were
parts of Ave pairs of pollinia adhering to the others in one
instance. This bee appeared to be doing good work, but
being a trifle under size, he had not the strength to extricate
himself.

This is an instance of lack of perfect adaptation of flower
and insect, for while the bees generally suffer no great in-
convenience in getting their su[)ply of food, some, and
especially the smaller kind.s of insects, are maltreated, with
no possible advantage to the flower, as they do not digest
the insects, nor their severed members, like Drosera and
Rhododendron viscosum. The above flower was A. Cornuti.
Either the many kinds of visiting insects did not have any
special taste for color, or else they were willing to sink
their tastes and run the risk of sundered limbs I

Apocyimm androsceviifolium has clusters of small, flesh-
colored bell flowers. The almost sessile anthers connive
over the pistil, meeting at the toy), but separated below by
a very narrow slit. Just opposite each slit at the base of
the corolla is a line tooth-like process which projects into
the slit, almost closing it but not quite, as there is left the
width of a hair between it and the anthers on each side.

Very small insects creeping in the bottom of the flower
get their legs caught in this narrow space and are held fast.
If they have strength to pull away from the tooth, the leg-
then goes into the slit, from which there is no escape. The
only advantage to the flower in this and the previous exam-
ple, would seem to he to punish intruders ; but the latter
flower has honey guides inviting all, if that is to be con-
strued as their object, which the writer does not allow ;
certainly there is no sign to the small fry to "keep out."

Azalea viscosa, Cnicus miiticui^, Nicotiana, several species
of /Silene, etc., which have viscid secretions on or near the
flower, are more considerate, as these give warning to the
ants not to trespass, but quite a number of small winged
insects are hopelessly glued down. In the case of Azalea
with its glands, the insects ma}^ perhaps be digested as in
Dromra, but it does not appear to be the case in the others.
The insects entrajiped are quite diminutive and the numbers
comparatively small, so it would seem that the loss of
nectar taken by them would be only a trifle, but a trifle
saved for more efficient fertilizers may in the long run be
of much importance to the flower. Who knows !

But an insect's life is subject to calamities ; what with
traps, pitfalls, smothering, drowning, starvation, assassina-
tion by other insects, and takings off by l)irds, etc., the
bees and the butterflies have their full share of trials and
tribulations, which are rarely noticed except by the close
observer of Nature, whose sympathy is often aroused.

AGENTS AFFECTING COLOR.

Dilute nitric acid changes purple or blue flowers and
leaves as violets, heliotrope, vinca, and leaf of coleus, etc.,
to a red color ; yellow flowers of oxalis and daffodil U)
brick-red ; other yellows as in tulip, lantana, alamanda,
cattleya citrina, to a blue-green.

It affects differently the different petals of Cytisus caua-
rieiisis, chano-in2' the "standard" to areen and the wings lo

98

brick-red. Some yellow flowers are not at all affected by
the acid. A yellow pansy quickly turns to a bright green,
followed by brick-red ; white flowers turn from pale to
chrome yellow.

Salsoda changes some purples to bright blue, others to
bluish ; certain whites to lemon yellow, others to yellowish-
brown ; some yellows, as in Lysimachia, to brick-red ; this
last evidently contains some red, for it may be detected
with a lens. The soda brings it to view.

Potash does not affect the yellow color of tulips, dande-
lion, or alamanda, but changes the lemon-yellow of oxalis
to blood-red ; it has no effect on the standard of Cytisus,
but changes its wings to a dull orange or amber color, and
Lysimachia to brick-red ; a purple crocus to green ; some
white flowers to lemon-yellow, others to bright red. A
scarlet pelargonium, it is said, changed by an alkali to
yellow, may be restored to its original color by an acid.

The purple vegetable dye, litmus, changes to blue with
an alkali.

Acids and alkalies are properties of plants ; various kinds
are free in the cell sap, and color is naturally influenced
accoi'ding to the proportions of each.

Ordinary atmosphere changes the color of a certain mush-
room from yellow to blue when fractured, and the yellow
sap of Baptisia becomes an indigo-blue by simply agitating
it in the air.

Ozone changes vegetable colors.

Certain minerals, as iron, change colors.

The action of the sun on the chlorophyl-green of stems
and leaves will often change that color to red or pur})le,
while the opposite side is unaffected

Too strong sunlight will often destroy color or cause it
to fade to white, while a proper amount deepens it.

Water will dissolve out certain purples, changing them
to white as in violets.

Field scabious (Knautia) of two shades of iHir})le, is said

99

to change to a bright yellow when ex[)o.sed to the .smoke ot
tobacco !

A proper amount of .sunlight is nece.ssary to evoke color.
Insects deepen and intensify the natural hue by the stimu-
lating effects of their friction.

Insects may change the color of an individual Hower bv
puncture.

Newbigin ascribes color in tlowers to the result of dimin-
ished vegetative power ; another has given an opinion that
change of color is due to "the struggle for life."

It is well known to botanists that flowers change color
by the mere process of drying, while in press.

A white Monotropa turns coal black, also yellow Ba})tisia :
some purple asters fade to white, and the white pap})us of
others turns red.

A rain drop or a spider, remaining for a few moments
on a petal of a purple morning-glory, will leave a red spot.

The sap in Borraginaceie, as forget-me-not, etc., is said
to be tirst strongly acid, l)ut that with age the acid disap-
pears. These flowers change from pink to blue.

According to Louis Prang, yellow^ can be ol)tained by a
mixture of red and green.

There are some indications that insects by their stinui-
lating influences may really evoke in a limited degree a
new or modifled tint by drawing some new element to the
surface, as suggested in Liriodendron.

Many flowers as a whole, or their honey guides, change
color at the time of fertilization. probal)ly by the plant
diverting its energy at that time to the setting of the seeds.
Different reliable gardeners have stated that cuttings from
a blue Hydrangea hortensis frequently })roduce at flrst pink
flowers. This may be accounted for by the new, rich soil
and better nourishing, for according to the same authority,
when the plant gets pot-bound and starved, its flo^vers are
apt to come blue !

100

ADDITIONAL REASONS WHY INSECTS DO NOT SELECT COLORS.

1. There are numerous species of flowers which have
never been known to change color, for example, dandelion
and buttercup.

2. A considerable number of different hues are devoid
of odor and nectar.

3. Flowers of the same species gro\vin<>; naturally in
different regions, are visited by (juite different insects.
According to Miiller himself there is a vast difference in
the number and kinds of insects visiting Thymus serpyllum,
whether in lowlands or on the Alps. "In lowlands there
are more flies ; in the Alps more butterflies."

4. Short-lipped insects would naturally seek o[)en or
diminutive flowers, but ordinarily there would be nothing to
prevent their entrance to tubular or bell-shaped flowers,
woY anything t(j prevent bees and butterflies from visiting
the former. Comparatively few flowers are specialized for
a single species of insects.

5. If insects have preferences and can change the color
of flowers by their partiality for their favorites and the
neglect of others, one would expect that the resultant of
the visits of a score or two of insects belonging to different
species would be a dull or neutral hue, while as a matter
of fact our flowers as a rule have remarkably pure and rich
colors, as seen in the daisy, buttercup), rose, etc., notwith-
standing that they may be visited l)y as many as tifty differ-
ent species of insects, or even more, including coleoptera,
hemiptera, lepido|)tera, and hunmiing-birds.

Miiller states that eighty-eight different species of insects
visit Cnicus arvensis, rose-i)urple, and ninety-three s})ecies
Taraxacum officinale, yellow. If these insects, each with
its special taste for color, can by their combined efforts
produce a pure yellow in one case, and a rich purple in
another, it must l)e because they were guided by Omnipo-
tence, and the I'esult is no less than a miracle !

101

(). The same s[)ecies of insects are the principal fertilizers
of flowers of totally different colors ; as bees in jSIelUotus
aJba, and Liguslrum, privet, both white ; Bapiisia lincloria
and Impalienf< fulva, both yellow; Trifolium and Veronica,
both purple l)ut of different grades ; jSaUx and Pteh'a, both
green.

In the Spring, the same bees being almost the only insects
in motion, visit one color of Crocus just as readily as an-
other, and honey bees visit the wholly green, pistillate
willow as promptly as the bright-flowered Crocus.

If it were urged that these divers colors have not changed
because they have never shown any tendency to change,
then it should be allowed that the larger portion of our
Flora is not at all affected as to color by insects !

7. Flowers change color in spite of their visiting insects.
The same bees visit the same flowers, continuing in the
same pathways to the nectaries, tracing thereby permanent
guiding lines made of the original color, which said bees
were supposed to especially fancy, but, while the guides
remain, and the bees continue as ])efore, the remaining
portion of the corolla changes to a strongly contrasting hue.
The bee selected nothing, hi.') remonstrances, if any, were
in vain, the flower for good and sufficient reasons just
simply changed color, and if he did not like it he could
leave it !

8. There is collateral evidence that color does not result
from insect or animal selection in the color of spathes,
involucres, sepals, and many pretty winter buds which
have no nectar nor odor and yet are found of every hue, in
numerous instances totally unlike that of the petals. In
such cases two different kinds of insects would be neces-
sary to accomi)lish the variegation ; are they each compet-
ing for the ascendancy, or one trying to confuse the other?
True foliage likewise changes color, as from red or pur})le
to green, or from green to yellow, red or [)urplc. Insects
ol)viou8ly are not concerned in these color changes, nor are

102

animals either. The same may he said of the stems.
Green hrier and sassafras are green, cornus and vihurnum
red, grape vine hrown, blackherry claret color, st)me l)irches
snow white, certain willows yellow, a host of shrubs gray,
a few goldenrod and adiantum stems black.

We are not aware of such a diversity of grazing animals
t)f individual tastes as would be sufficient to meet the re-
quirements of the infinite variety of hues displayed in stems
and foliage. Donkeys and goats do not run wild in New
England, and rabbits, deer, bears and bison are not {\?sthetic
in their tastes. Nor do we believe that there exists a suffi-
cient number of burrowing and rooting animals (earth
worms are said to be eyeless) to account for the numberless
varieties of tint in roots, bulbs, etc.

The common garden radish is of a pure red, carrots
orange color, parsnips pale yellow, beets dark red-purple,
turnips, some white with purple tops, others yellow Avith
purple bases, onions run from a porcelain white through
every color to purple or blue, and the whole range of color
is also found in the roots of our native wild plants.

9. If blue flowers are especially attractive to "azure-
loving bees," one might naturally expect in view of the
numerous species of Apidoi, that blue would be the pervad-
ing color, or at least a very prominent one, but it is a well
known fact that l)lue flowers are comparatively rare. I do
not recall any tree or shrul) in New England l>earing a blue
flower. Moreover, it is obvious that many blue flowers are
wholly inde})endent of bees. We have already alluded to
Larkspur as not being adapted to bees on account of its
long spur. The widespread blue flower of Clematis, Jack-
manni, is not nectariferous, and is only occasionally visited
for its pollen, while several other species of Clematis of a
blue color, have tubes so long and narrow that both bumble
and honey bees are excluded from the honey, unless they
bite holes from the outside, as C. lubulom, C. Uavidicma,
etc. These flowers of our gardens are exotics, but it is

103

clear that they were never developed by nor intended for
bees ; the same is true of garden phlox and verbena.

Even among ))lue flowers there is a marked difference in
the visits of liumble and honey bees ; honey bees enjoy,
heliotrope, bum1)le bees do not visit it; bumble bees delight
in pickerel-weed, honey bees are not attracted ; bumble bees
only, visit pansy ; I have not seen honey bees on Anchusa, etc.

Of two rows of Snapdragon, one yellow and one purple,
each bee followed down his own row to the end without
crossing to the other which was equally near, also in a bed
of Crocus each bee generally takes a single color, and when
l)ursuing yellow colors will fly directly over and across blue
flowers to find the object of their search, which outside of
flower gardens means a particular species of flower, to which
they are guided by the color. When certain species are
scarce they are not so methodical. This clearly indicates
that if a particular flower satisfies their gustatory tastes they
do not abandon it simply for one which pleases the eye.

Bees take their repasts mostly in courses. It will prob-
ably be found that they know the hours at which the differ-
ent species blow, and are then promptly on hand; they
will take the roses in the morning, and the honeysuckles
approaching sundown ; crepuscular moths appear with won-
derful promptness as soon as Oenothera expands.

In changeable flowers they take the freshest, whatever
the color then may l)e ; and they often are obliged to dis-
criminate against purple and blue !

Bees are able to distinguish between different sorts of
blossoms of the same color, but a difference in hue is of
great assistance in finding the kind of flower they want.

We see changes going on in the individual flower under
our very eyes, without the aid of insect selection : yellow
Pansy to blue, white Lonicera to yellow, etc. What takes
place in an individual flower in a few hours, we can well
conceive might require under other circumstances many
generations to effect, and the change would just as surely

104

be made without insect selection in the latter case as in the
former.

When the more recent leaves of a Croton change from
yellow to scarlet and the older green leaves at the same time
change to a red-chocolate color, no insects whatever are
concerned in these color changes, and yet the new colors
are as distinct and well defined as those of a flower. No
insects are concerned in the colors of bracts, leaves, stems,
petioles and roots ; why in the parts of a flower?

10. There are a plenty of existing flowers of a dull hue
owing to crude combinations of pigments, as perhaps green
and purple. Why have the bees not selected the purple
thousands of years ago? Simply because they have no
control over the acids, alkalies or other chemical agents in
the sap. Their utmost ability is to perpetuate what the
constitution of the plant creates.

The white w^ater-lily (jSly7npha>a) in south-eastern Massa-
chusetts is slightly inclined to a pink color, discoverable on
the borders of the sepals. On Cape Cod there is a variety
wholly of a beautiful pink tint. Does anyone believe that
this tint was produced by the selective agency of some
insect peculiar to Cape Cod? The idea would probably
strike the average mind as absurd.

In brief. Mailer's idea is that insects have changed the
form and the color of flowers, which is equivalent to saying
that insects have created the different s})ecies of [)lants.
The result of this doctrine logically carried out to its final
analysis would be that, as species affect the fruits, all the
various kinds of fruits, as plums and cherries, apples and
pears, oranges and pineapples, etc., were originated practi-
cally by bees and butterflies ! I do not think that with our
present knowledge we are quite ready to accord to them so
much power and influence over the floral world.

W^ith all deference, therefore, to Muller's opinion regard-
ing Gen/iana, we are forced to believe that he was mistaken
regarding the cause of a different color in a different
species.

105

CONTENTS.

Page.

Sequence of Colors, . . . . , . 5

Primary Colors, ....... 17

Colors of the Ranunculacefe, .... 20

Theory of Grant Allen, ..... 25

Changes of Color in the Individual Flower, . 29

Yellow Flowers, where found, .... 33

Color Studies, . . . . . . . 37

Striking Contrasts of Color in Flower and Leaf, . 41

Green to White, etc., ...... 45

Sap, Pollen, and Pigments, ..... 47

Conclusions on Color Sequence, .... 54

Honey Guides : . . . . . . . 55

Tropteolum,

Spanish Iris,

Poppy,

Tulip Tree, etc.

Night Bloomers, 72

Variegation of Leaves and Seeds, ... 74

Do Insects Select Colors? 76

Principal Visitors of Certain Flowers, ... 78

Cruelty of Flowers, ...... 95

Conclusions as to Selections bv Insects, . . 104

OBSERVATIONS

COLORS OF LEAVES

PRKCEDEI) in A

SUPPLEMENT TO

Observations oi| the Colors of Flowers,

15 V

E. WILLIAMS HERUEY.

NEW BEDFORD:

E. Anthony ^V; Sons. Incori).. Piuntkus

('()l)yrif;ht, HUH, I>y E. W

SUPPLEMENT

TO

OBSERVATIONS

ON

THE COLORS OF FLOWERS.

Among various notices of my Observations on the Colors
of Floivers, certain criticisms which appeared in The Amer-
icrrn Naturalist and The Gardeners' Chronicle, may conven-
iently, and, as it seems to the writer, very properly he
referred to in this supplement to the same.

The Gardeners' Chronicle of Feh. 10, 1900, while giving
on the whole an appreciative notice of my pamphlet, sug-
gests, very courteously, that on the subject of honey guides
I might have "unconsciously absorbed and api)ropriated '"
Mr. Henslow's theory.

But as a matter of fact 1 had not read or heard of Prof.
Henslow's Orir/tn of Florcd Structures until 1 had formed
my opinions from my own observations upon the genesis of
honey guides, and the "copy" was ready for the printer.
Moreover it would seem that The Gardeners' Chronicle
wholly misapprehended my views u})on the subject, although
special pains were taken to present them clearly. As my
ideas were totally different from those of Prof. Henslow,
both his and mine will now be re-stated, l)ut in different
phraseology from the originals.

Prof. Henslow's method may be understood by sup[)osing
bees or other insects, in pursuit of nectar, to alight habitu-
ally, season after season, ui)on a j)articular petal ()f, tor

instance, a white flower until tinally a yellow spot be pro-
duced upon the white foundation, as a result of the con-
tinued insect irritation. The yellow spot would be called a
honey guide.

Prof. Henslow's own words are: ''The spots have been
determined by insects " ; " they are srm.plij the direct results
of the insects''' ; "one result of a more localized flow of
nutriment"; "the insect visitors induced the flower to i)aint
the petal with a golden streak."

The writer's method supposes, for instance, a yellow
flower to be visited in like manner, but the irritating effects
of the insects result simply in deepening the existing color in
the one irritated spot but not otherwise changing it, as in
the preceding case. Thus far we have a flower still of a
single color but with a part of one petal usually, of a some-
what deeper shade. Numerous cultivated flowers are to be
met with, in practically this condition, without any more
striking feature. To produce a strongly contrasting spot,
marking, or entire petal, as is more commonly seen in
honey guides, something more is required, which is not
within the power of insects to effect, namely, a general
change of the basal color, except in the circumscribed and
irritated spot before mentioned n-hlch does not change.
This general and often profound transformation is not lim-
ited to any one color, but a very common change is from
yellow to white. Our entire flower would therefore have
changed from yellow to white except in one spot, and that
spot is yellow because it is part of the original color. The
honey guide thus formed has the old color ; Prof. Henslow's
honey guide has a neic color. Our flower has a new foun-
dation color, while that of Prof. Henslow has the old, origincd
one. The methods it will be oljserved are as unlike as could
well be imagined: there is absolutely nothing in connnon
and yet we are told that our method is appropriated from
Prof. Henslow !

Prof. Henslow remarks farther reijardino- honev liuidos :

5

"Those peculiar and special displays of bright tints distrib-
uted in spots and streaks in certain and definite places only,
have been called 'guides' and 'path- finders,' as they invar i-
ahly lead to the nectaries."

In this statement. Prof. Henslow falls into the same error
as several other writers, the fallacy of which is pointed out
on pp. 67, 70, and 74 of \wy Observations.

Joseph Y. Bergen in Tlie American Nalnralist for No-
veml)er, 1.S9U, referring to the comparison of the effects of
mechanical irritation on animal and vegetable tissues on
p. 58 of my Observations on the Colors of Floiners, viz. :
that the bees stinudated the flower to send more pigment to
the honey guide, as friction or a pinch would bring more
blood to the cheek, observes as follows :

"Even the lay reader will find little difficulty in judging
of the value of the analogy between the development of
pigment in cells of petals and the response of human arte-
ries to stimulation transmitted from the central nervous
system."

My statement is perfectly plain, viz., that the additional
pigment which made more depth of color in the flower
mentioned was sent to that point the same as the blood was
brought. But the critic has the faculty of wresting the
language of the text to suit his own ideas, as is seen in his
accompanying conniient. It seems inexcusable that he
should represent me as saying that the pigment was devel-
oped in the cells when I distinctly say that the pigment was
sent to the cells. Furthermore the introduction of the sub-
ject of the Central Nervous System was purely gratuitous,
as was the absurd analogy which he himself invents, and
upon which he asks judgment.

The cause of the redness of the skin from friction is
solely the relaxation of the coats of the blood-vessels,
allowing the ordinary flow of' blood to dilate them.

We apprehend that the cause of more color in the honey
ouide, to which reference was made, was that the irritation

6

and pressure of bees and of other insects upon the spot
caused a relaxation of the cell-walls, which became thereby
more permeable, and that more colored sap was attracted to
the stimulated spot. There was in the process no extra
heart pressure on the one hand, nor more root pressure on
the other. The flower being of one uniform color as stated,
the stimulus at one point would naturally draw some of the
colored sap from the neighboring- cells to that point, pre-
cisely as more red blood is drawn to an irritated spot simpl}"
because the caliber of the vessels is enlarged.

The analogy therefore is, that local friction brings color
from the immediately surrounding parts, and that this addi-
tional color in l)oth plant and animal is confined exactly to
the limited s})ace which is irritated. Whether my analogy
(not his) is to be regarded as yalual)le or worthless will not
make the slightest difference to the main facts given regard-
ing the origin of the honey guide, which the critic com-
pletely ignores.

I may be excused in extending these remarks somewhat
upon the subjects introduced by the reviewer. Undoubt-
edly the same irritation of the insects might, under some
circumstances, attract colorless sap, laden nevertheless with
pigmental elements, from which the chemical character of
the cell contents would develop color. If for instance the
cell contents of the petals should happen to be of a slightly
acid character, a certain color might develo[), while at the
same time if some of the same general sap should enter the
cells of another part or organ, as the stamens, whose cells
might chance to be slightly alkaline on account of their
special functions, these would l)e more than likely to de-
velop a different color. AVhether a flower is to ])e red or
blue may depend upon which of these elements prevail, and
how long it i)revails, for cell contents vary, and color varies
with them. The sap in Borracunacem is first strongly
acid, but as the flower develops, the acid disappears. The
flowers commonly change from red to blue.

But whether the [>ii>ment was actually brought to or
developed in the petals' cells, would really make but little
difference to the analogy, inasmuch as there would be an
actual movement of sap in either case. In our example
however not even this possibility was suggested.

Mr. Bergen alludes to "the development of color in cells
of plants," a sul)ject not mentioned in my text, but in pass-
ing I will point out some resemblances in plant and animal
as to color. Cells, of course, and the develo})ment of color
in them, are not i)eculiarities of plants : the bodies of ani-
mals as well, are com[)()sed of cells, and even the i)lood
vessels are only lengthened cells. As pure and brilliant
colors are derived from red blood as from j^olorless saj).
These colors are more conniion among those lower animals
which in sim})licity of organization are nearer to the vege-
table. Many sjjecies of worms have colors of much beauty ;
the criisfacece are remarkable for the l)rilliancy of their
coloring ; reptiles are highly colored ; there are green frogs,
black and yellow salamanders, snakes and lizards of innu-
merable tints : "in insects the colors of the two stagi^s larval
and adult are often strongly contrasted, and such beauty of
pattern and design cannot be found in the tioral kingdom " :
birds and fishes have both structural and pigmental colors,
while mammals are rarely remarkable for brilliant pigments ;
the different races of man have quite a range of color,
white (?), red, brown, and black, and disease of the liver
may produce a yellow color, but the principal display of
color is contined to the hair and the visual organs, which
contain real pigments liable to change of tints as in flowers,
and the internal organs have various tints. These colors
are separated in the animals from the nourishing constitu-
ents of the blood, by dialysis, chemical action, and to some
extent by the effect of solar light closely similar to the de-
velopment of color in vegetable cells from the colorless sap.
Moreover some lipochromes identical in character are pro-
duced in the animal and the vegetal)le I

8

When the color of the Hower and that of the sap are
.similar the resemblances are apparently nearer. Many*
plants have laticiferous vessels in which the latex circulates
as freely as the blood in the veins. The latex varies in
color in different plants, beinu" white, red, purple, or yellow,
etc., and is frequently of the same tint as the flower, as
seen in Celandine, Tradescantia Virg'inica, Sansfuinaria
Canadensis, Apocynum, etc. ; in Sanguinaria the flower is
usually white, but sometimes rose color, and in Apocynum
the white latex is often tinged like the flower, being white
and rose. The latex and colored saj) is apt to vary in color
in its course, likewise blood varies from scarlet or crimson
to dark purple.

In a cyclamen with purplish })etioles and scapes, the
purple vascular bundles surrounding the central part or pith
were traced uninterruptedly from root or bulb to the red
purple flower, a cross section of the scape displaying a
purple ring, and numerous pur})le dots which were the ends
of scattered bundles. Similar colored bundles are to be
seen in Primula sinensis, and P. stellata, a cross section
showing a crescent ; the l)undles run to the ribs of the
leaves, also colored, and u}) to the flower which may be of
a different color, or they may be broken, by an uncolored
gap, from the flower. All the sap of a beet root is red, and
it sup})lies the red petioles and the ribs of the leaves. But
of course sap is usually colorless.

The gratuitous introducti(fn by Mr. Bergen oi the subject
of nerves, was obviously for the })urpose of showing an
immense difference l)etween animal and vegetable tissue,
and the impossibility of any analogy on that account. He
is very positive that the redness i)roduced by friction is due
to "stimulation transmitted from the central nervous sys-
tem" ; but there are serious doubts as to that on the part of
the writer. Whither the stimulation is transmitted he does
not say, and one is obliged to enquire whether he means
that it is sent to the heart, or directly to the reddened skin.

9

That the heart is not stiumlated is cleai- from the fact that
the pulse is not api)reciably (juickened, and that the effect
is not general, hut local. Some facts may he of interest
concerning the effect of stimulus upon the blood vessels,
and the effect of a suspension of all stimulus.

''When a person blushes the action of the nerves upon the
vessels of the cheek is temporarily suspeiided, causing them
to dilate." (Huxley and Youmans).

"AVhen })ure alcohol is applied to the skin, after the tirst
effect of cold caused by its evaporation has [)assed off, the
part becomes red from temporary pared >/nis of the bJood
vessels, causing them to dilate." (Prof. H. N. Martin).

"If the sympathetic nerve in the neck of a rabbit be di-
vided, a vascular congestion of all parts of the head on the
corresponding side immediately follows. The vessels of the
ear become turgid with blood." (Prof. J. C. Dalton).

"Terror causes the skin to grow cold, and the face to
appear ^ja/e and pinched, the supply of blood to the skin is
greatly diminished in consequence of an excessive stliimla-
tion of the nerves of the small arteries which causes them to
contract.'' (Huxley and Youmans).

It will be observed that in each of the tirst three examples
preceding the skin becomes reddened when there is no stim-
ulation ivhatever from the central nervous system, and in the
case of the rabbit all communication was entirely cut of from
ithy section of the nerve. The effect therefore oi a shs-
jjension of stimulation, is relaxation of the vessels, dilatation,
and redness ! If the blood vessels are really stimulaied by
an influence transmitted from the central nervous system as
Mr. Bergen implies, then it appears that the same effect is
produced when they are stinuilated, as when they are not !

The subject of nerves is a very difficult one, and notwith-
sbanding the increase of knowledge on the subject in recent
years, it is still far from being well understood. It is
known however that there are vaso-constrictor nerves which
contract the vessels, when stimulated, and vaso-dilator

10

nerves in certain parts of the ))ody which are said to dilate
the same when stimulated, and in certain nerves both kinds
of libers have been discovered. Several authors state how-
ever, that the action of the dilator neives is of an inhibitor^/
character.

One writes that "inhibition is connected with tlie effect
of two sets of impulses upon the responding cell, and the
two stimuli must tend to excite different reactions." There
is little doubt from a consideration of the subject but that
the cause of the relaxation is a removal of the normal tonic
action by inhibition of the activity of the constrictor nerve
centers. The l^lood vessels themselves would in that case be
temporarily without stimulation, and practically in the same
condition as when a nerve is paralyzed or when it is severed,
as in the examples above stated. Inhibition, which checks
or restrains the constrictor nerve centers from sending' out
their ordinary impulses, is a totally different thing from
stimulating them to action. From the foregoing facts it
would seem therefore, that while the relaxation is due to the
action of the nerves, the action is a negative one, and that
no stiviidation is transmitted from the central ner\'ous system
to the coat of the blood vessels I

But, when the nerves resunie their activity and stimulate
the muscular fibers, those libers again contract, and the
result is, smaller tubes, and lessened color I

It should not ])e overlooked, while referring to the action
of nerves, that muscle has in itself, wholly independent of
the nerves, the properties of contractility, irritability, and
conductivity, and that therefore muscular liber can be ex-
cited to action not only indirectly by the nerves, but by
application of the stimulus directly to the nmscle. Inas-
much as vegetable tissue has precisely these same })roperties
it can well be imagined that plants and animals have many
functions of a similar nature : indeed many of the lower
animals have no nerves at all, and so nearly resemble vege-
table tissue that the two classes are with difficulty distin-

11

guished, and yet the fonner havo the power of locomotion
and other niovenient.s.

The human tlesh is competent to heal slight injuries bv its
vital action ; the same is equally true of vegetable tissue
which has no nerves and yet can heal greater injuries. Irri-
tation of the skin has been known to produce various kinds
of tumors in man and beast, especially in horses and dogs;
various outgrowths on Howers, roots and stems result from
the same cause.

The hair of a horse where the skin is galled by the har-
ness freiiuently changes from l)rown or black to white ; the
color of a leaf may be altered by an injury of the stem or
petiole.

To show more fully the analogy existing between nerve
in animal and irritability in [)lant, we submit the following
quotations :

"The property of conductivity (of transmitting a condi-
tion of activity aroused in one part by a stimulus to any
other portion) is exhibited l)v a vast variety of forms of
cell-protoplasm, and bi/ plants as well as animals.

"In the case of plant-cells and in certain forms of muscle-
cells about which there is a more or less detinite wall or
sheath, there are little bridges of protoplasm binding the
cells together." (Am. Text-Book Physiology).

There is an irritability of muscular tiber distinct from irri-
tal)ility of nerve. This independence of irritability is
proved in many ways :

"By the use of the drug curare we are enabled to prevent
the nerve impulse from reaching the nuiscle, and when we
have done this we tind that the muscle is still able to re-
spond to direct excitation, with all forms of irritants, viz.,
electrical, mechanical^ thermal, and chemical." (Am. Text
Book Physiology ) .

"Some parts of muscles, such as the lower end of the
sartorius, and many muscular structures which have no

12

nerve terniiuals in them respond energetically to all kinds
of muscle stimuli.'" (Prof. J. C. Dalton).

"There are some substances which act as stimuli when
applied directly to the muscle, but have no such effect upon
the nerves, viz., ammonia." (Prof. J. C. Dalton).

" In some of the lower animals there are simple forms of
contractile tissue in which nerves cannot be discovered and
which are irritable.'' (Am. Text-Book Physiology).

"The scientific biologist recognizes the fundamental like-
ness in the structure and functions of plants and animals.
Plants feed, breathe, and reproduce exactly as do animals ;
respiration takes place in all plants precisely as in animals.

"Protoplasm, the physical basis of life, is in each and is
much alike in each. The cause of the movements of plants
is largely due to the direct effect of light upon the sensitive
2)rotoplasm of the cells of the motile i)arts." (D. FI. Camp-
bell, Ph. D.)

"Plants are exhausted by over exercise and require rest,
and like animals they are lulled and put to sleep by chloro-
form and narcotics ! The faculty of responding to external
irritation by internal movements and change of form belongs
to cells and holds good in the vegetable as in the animal
kingdom. All M. Cohn's experiments prove that in Mimosa
pudica, which is highly sensible to the action of light, heat,
electricity and touch, the propagation of the external excite-
ment proceeds in the same mode as in animals." (Somer-
ville Molecular and 'Microscopic Science).

"Mechanical irritation of the glands of Drosera rotundi-
folia by insects causes secretions." (Prof. Goodale).

"The three delicate hairs on each face of the trap and the
median line of Dioncea muscipida are exceedingly sensitive,
the lightest touch upon one of the hairs will cause the valves
to close instantly." (Prof. Goodale).

"In Drose^^a the secretions of the hitherto neutral glands
become acid in consequence of the stinmltis of insects. On
stinnilation of Dionoea, electric currents arise in the leaves,"

13

etc. Tendrils are distinguished (from twining- plants) l»y
being in a high degree irritable especially to nntiact ici//> a
solid body.

" If a leaf of Mimosa be stimulated by means of the hot
focus of a burning glass, not only that leaf folds together,
but the stinuilation extends, till all the leaves of that branch
have the movement" !

"The aiillions of stomata in an average size leaf simulta-
neously open when the sun shines on the leaf, and simulta-
neously close when it l)ecome8 shaded."

The leaves of many plants are so sensitive as to close in
slightly cloudy weather, as oxalis and anagallis. Numerous
flowers close at night and open in the mornino-, etc.

"We have no necessity to refer to the physiology of
nerves in order to obtain greater clearness as to the phe-
nomenon of irritability in plants; it will perhaps on the
contrary, eventually result that we shall obtain from the
processes of irriiabiliff/ in plants data for the explanation, of
the pJiysioloyij of nerves."' (J. von Sachs).

"The irritation set up by insects themselves is one of the
most potent causes of a flow of sa[) to certain definite places
which encourages local growths. The effects under mechan-
ical irritations and strains, of nutritive matters of the same
kind, of poisonous substances, of electricity, etc., all show
that the bond which unites the animal and vegetable king-
doms is of one and the same nature." ( Prof. (ieo. Henslow) .

"When muscular til)ers are touched by a pointed instru-
ment they exhibit contraction even after they have been
detached from the body, provided that too long a period of
time has not elapsed. At one time it was supposed that the
contraction of a muscular fiber depends so completely upon
the agency of the nervous system that it might be consid-
ered as the direct function thereof ; but a more critical
examination of the circumstances of the shortening of the
til)er cells, shows that it })ossesses many features in common

14

with the same contraction of the cells of plants which have
no nervous system." (Prof. J. W. Draper).

"It is impossible not to be struck with the resemblance
between the foregoing movements of plants (sensitive plants)
and man}' of the actions performed unconsciously by the
lower animals, etc. Yet plants do not of course possess
nerves, or a central nervous system ; and w^e may infer that,
with animals, sucJi f^lnicfvres serve only for (he more perfect
transmission and for the more complete intercommunication
of the several parts." (Darwin).

"In the last instance (iinal analysis) indeed, I might say,
animal and vegetable life must of necessity agree in all
essential points, indudin;/ the phenomena of irritability "\'
(J. von Sachs).

In addition to Mr. Bergen's analogy before mentioned,
which he would have the reader believe to be mine, he gives
a mangled and distorted account of my explanation of the
origin of the honey guide by quoting one half of the same
only, omitting a very im[)ortant part, which cannot fail to
convey a false impression. While seeking for Mr. Bergen's
views upon the color of autumn leaves (to be referred to
later) in his small volume of Elements of Botany, the
writer's attention was arrested by two remarkable state-
ments : —

"Old leaves are found to be loaded with mineral matter
left behind as the sap drawn up from the roots is evaporated
through the stomata." p. 117.

Mr. Bergen appears to be ol)livious of the fact that sap
may contain many ingredients other than mineral matters
which are not evaporated, and that these substances may
occur even in the roots of plants, as sugar in beets, sweet
potatoes, etc., albumen in the juices of carrots, turnips, etc.,
gluten in the juices of turnips, starch in sago, tapioca and
arrowroot, etc., all of which go to build up the stems,
leaves, flowers and seeds and do not go off into the air I

Mr. Bergen evidently confounds sap with water.

15

Referring to stomata, Mr. Bergen writes: "To some
extent they regulate the rapidity of trans[)iration, opening
77iO)'e irulel// in damp weather, and closing in dry weather,"

Vine^' Phyxiologij of PJanU however states, that Molden-
hauer supported by Amici, von Mohl, and Unger, find that
the stomata are chimed on rainy days and dewy nights, and
are open when the sun is shining upon the leaves I

Such statements as the foregoing indicate that an hyper-
critical reviewer may himself be open to criticism.

TRANSFERABILITY OF COLOR.

Most of the coloring matters in cells can be dissolved out
by })ure alcohol. The veins of petals and leaves are speed-
ily tinged with colored solutions absorbed through the
stems ; the cellular tissue may also be tinged somewhat, but
much more tardily. A white petaled Cyclamen with a
crimson projecting central part or eye, placed in alcohol
changed to white at the center, and after the alcohol had
evaporated, it changed back to crimson, but a portion of the
red pigment, evidently dissolved out of the cells, penetrated
and tinged the veins of the reflected wdiite i)etals nearly to
the tips.

The dark pur[)le eye of an Ixia, after remaining in a
tumbler of water for a couple of days entirely disappeared,
but the coloring matter reappeared in the veins and cellular
tissue of the white outer })ortion of the })etals. The yellow
[)igment in flowers, not l^eing dissolved in the cell sap, but
associated with granular substances in the cells, is not easily
transferred through the cell walls.

In order to test the transferability of coloring matter by
some mechanical device roughly approaching the irritations
of insects, experiments w'ere made by brushing the colored
surface of the petal with a small camel's-hair brush such as
is used in water colors.

Several blue violets taken from conservatories in winter
were selected, which, owing to cloudy weather [)erhaps,

16

were iiiipei'fectly colored, and had many small green specks
on the petals. When these specks, or diminutive spots,
were brushed across for a few moments they disappeared,
being lost in the general blue color, and blue veins a}y3eared
in the whitish area at the base.

Similar effects were produced on the white specks in
sweet william, and on the white halos surrounding the
purple dots of foxglove.

One of the maroon spots on a yellow pansy was brushed
lightly for 10 or 15 minutes, when numerous fine purple
veins ai)peared at the outer border, while in the interme-
diate space few were seen, they being more deeply imbedded
in the cellular tissue. A white pansy with blue honey
guides gave lietter results. A white variety of Martha
Washington geranium, with dark maroon honey guides on
the two upper petals, when treated in the same manner,
became covered with a network of dark red lines, while the
parenchyma itself was largely tinged with the same color.
The dilute coloring matter must have been conveyed by
ca})illarity through the vascular bundles, and diffused
through the cellular tissue by osmosis, as no pressure was
brought to bear beyond the area of the honey guide. When
two different colors meet, as the basal color of a petal with
the honey guide, an intermediate color is sometimes pro-
duced naturally, by the blending of the two as in water
colors, which would also seem to imply that there was a
diffusion and commingling of the contents of the contiguous
cells.

An experiment was made to ascertain if the color of a
flower could ))e changed or in any way modified by mechan-
ical irritation on the surface of its petals. The writer
noticing that Primida Sinensis was apt to change color
with age, selected several varieties of the flower and pro-
ceeded to manipulate them w'ith a camel's-hair brush. A
deep crimson variety with a green eye was chosen for the
experiment. Holding a flower while still attached to the

17

plant by the left hand, it was brushed with the right liohtly
and rapidly with the soft haired brush, with a circular move-
ment arcHind, but outside the green eye. After a few min-
utes' brushing a slight alteration in the hue was detected ;
continuing the process, soon a new and totally different
color began to declare itself. The brushing was continued
for fifteen minutes in all, when there was displayed a per-
manent circular band of a violet color around the green
center, the untouched outer portion retaining its original
crimson hue.

Another blossom freshly expanded was tested. It was
brushed for a few minutes only, when it became of a pale
violet color, and then, it being a late hour in the evening,
the plant was set aside till morning. In the morning it was
found that this pale violet color had faded to a clear white !

This Chinese primrose plant therefore had beside the
regular crimson flowers, one with a deep violet circular
band, and another with a white band !

The results were not only interesting, but decidedly sur-
prising ! Some time later a variety was procured with a
rich purple eye, otherwise of a similar crimson color. The
pur[)le eye in this was experimented upon ; with a few
strokes of the brush, lasting but a moment or two, every
particle of the purple color disappeared as if by magic and
was replaced by green. Similar results were obtained from
other varieties, some more successful than others, iind de-
pending somewhat upon the age and condition of the flower.

Referring to these colors produced by friction upon the
Chinese primrose, it would seem that some natural examples
of red colored flowers with blue or white spots might be
found in which insects induced the spots. In fact such are
occasionally to be found. Among the primroses themselves,
whitish areas are often seen near the base of the corolla,
and also in other blue flowers.

Viola cucullata, which at first is usually of a reddish-purple
hue, when growing in wet woods has generally a bluish or

18

violet spot more or less distinct at the base of each petal ;
on uplands the spots are less common, and the flower is
variable in this respect. A colony of V. cucullata was
found by the writer in a moist field, all the flowers of which
had white spots instead of blue ! These two colors, the
same as were produced artificially, were doubtless caused
by insects.

Those artificially produced were not induced evidently
by any additional nutriment, or localization of new color
material, but rather by some chemical change in the sap
already present in the cells, caused by the stimulating effects
of the friction.

A red variet}^ of poppy with large, round, white spots
instead of black, suggests a similar origin.

Insects often indirectly change a color by changing the
character of the vegetable tissue. This is apparent in galls
and other excrescences produced by bites, punctures, or
stings. A small worm hatched from a minute egg in the
stem of Solidago rugosa, will cause by its irritation an en-
largement of the stem. This swollen part, if exposed to
the sun, generally becomes of a purple color, whereas the
rest of the stem and leaves is perfectly green; in like
manner red nodules are caused on the ribs of the green
leaves, and tendrils of the wild grape vine, and red excres-
cences on the oak and wild cherry, and white balls on oak
leaves, etc. The puncture of an insect may change the
green calyx of Oenothera biennis also to red. The change
of color in these examples follows some change in the char-
acter of the tissue and a change in form. Changes in the
forms of flowers and many peculiar outgrowths from the
petals are also caused by insect agency.

If one sees the lip of an orchid, two, four or six times as
broad as the other segments and knows that that lip is
habitually used as an alighting place for the insects visiting-
it, he hardly needs a demonstration to prove that its enlarged
size was due to their presence.

19

A new outgrowth, caused by insects, and but sliirhtly
moditied in tissue, appears therefore to favor a change of
color in stems and leaves, and the same tendency is seen in
the petals of flowers. The two upper lobes of Rhododen-
dron (Azalea) under a lens will be found to be obscurely
wrinkled with slight elevations and de[)ressions. The dots
or hyphen-like marks occur on the elevations. A slight
change of tissue, and their prominence cause a concentration
of coloring matter in the dots, which are always of a deeper
shade than the basal color. The beard on the petals of
certain species of Iris is doubtless caused in a similar
manner by the fretting effects of insects such as bees with
sharp clinging claws on a very sensitive surface. It is not
improbable that the yellow color of the beard on an other-
wise blue flower, is due to the fact that the beard is a com-
paratively recent outgrowth of tissue of a different character
as regards the cells, and probably of somewhat different
functions Calopogon jjuIcheUus has a crest-like outgrowth
on the lip, and the teeth-like processes have different colors,
white, lemon-yellow, orange and purj)le, according to their
varying lengths and forms.

The central, yellow eye in Pansies has papilhe several
times as long as in other parts of the flower.

A white variety of Weigelia rosea with age changes to
pink or rose color, also the two yellow, narrow streaks in
the throat, but the area within and about these streaks,
being the })art stimulated the most by the bees, becomes a
deeper shade than the rest. Catalpa spectosa has two yellow
lines and numerous purple dots as honey guides. With age
the white ground upon which these markings occur changes
to purple, but no other part of the white corolla changes.
The influence of the bees is thus plainly indicated in deep-
ening color, or producing color where there was none pre-
viously. In the morning glory, a rain drop or an insect
resting upon a petal, while the sun is shining, is apt to
change the color beneath it.

•20

The colors which are quickly changed, either naturally or
artificially, are not likely to be lasting, and the latter gene-
rally merely anticipate what would naturally take place in
due time later.

The colors of new outgrowths appear to be developed
with, and as a part of the new tissue, as in the beards on
petals, etc., but such outgrowths are rather of an excep-
tional character, whereas spots, streaks, and dots on the
plain surface of a petal are of very common occurrence, and
are ordinarily remnants of the original color, and have
become hereditary. The scarlet zone in the Kaiser Crown
tulip, referred to on p. 64 of my Observations, ^vas thought
to have been developed directly from a different base-color,
but from an examination of additional specimens, it is found
that this scarlet zone, like the ordinary honey guide, is also
a remnant of a color which once extended upward to the tip
of the ])etals ; it therefore affords an additional proof of
the correctness of the method above described.

We have seen by the foregoing statements that insects
indirectly change the color of stems and leaves as in galls,
etc., and that of flowers by causing new outgrowths as hairs,
beards, ridges, crowns, nectaries, etc., and also directly by
stimulating a plain surface as in the Chinese primrose ; that
their irritations on the petals tend to transfer and diffuse the
coloring matter dissolved in the cell-sap, or to concentrate
it into spots and streaks according to circumstances — di-
rectly, or indirectly by bringing more colorless sap to be
colored within the cells, or by so affecting the tissue that
when a flower changes color by age, the stimulated part only
may change, or if the change is general, then the stimulated
part acquires a deeper color.

OBSERVATIONS

ON

THE COLORS OF LEAVES.

Every one is familiar with the colors of autumn leaves,
but comparatively few have })aid attention to the more
(juiet, yet beautiful, hues of the young and tender leaves
of many trees and shrubs in springtime.

Perhaps one-half of the different species display, in the
unfolding leaf, colors other than green, such as silv-er-gray,
pink, red, purple brown, or bronze green, with a very few
of a yellow or white color. These colors are for the most
l)art of a transitory character, lasting for a few weeks only,
when they are gradually replaced hy the more permanent
and all prevailing chlorophyll green. They are not peculiar
to spring, but ai)pear on all recent leaves of the same i)Iants
throughout the growing season until fall.

It has been thought that these temporary colors may serve
as a screen to protect the young chlorophyll granules from
the too warm rays of the sun. In certain plants as the
purple beech, purple barberry, etc., the color persists until
the cool fall weather and then disappears. While these
temporary hues of spring foliage resemble in some respects
the autumnal hues, they are evidently of a different chem-
ical composition, for the latter result from the disintegra-
tion of the chlorophyll, while the former appear to be due
to additions to the cell contents ; yet many of the same
trees and shrubs conspicuous for color in the spring, after
their foliage has completed its summer work of assimilation,
array themselves again in very similar but nuich brighter

22

hues than before, such as the Oaks, Maples, Poplars, Sassa-
fras, Yaccinium, Prunus, Viburnum, Rubus, Ampelopsis,
Amelanchier, Leucothoe, Clethra, Vitis, Salix, etc. A few
of the preceding are purplish in spring and yellow in fall.

As tending to strengthen the view that spring colors are
screen colors, is the fact that many plants display these
colors only in the sun, but in the shade they are entirely
green, as Vaccinium vacillans and V. Pennsylvanicum
(blueberry).

In addition to these spring colors, the leaves of numerous
species of a semi-evergreen character which have weathered
the winter months, but changed to brighter hues where
exposed to the sun, still linger on their stems, as Black-
berry, Rumex, Strawberry, Cranberry. Boxberry (Gaul-
theria), Smilax, Chimaphila, Pyrola, Mahonia, Lonicera,
Kalmia augustifolia, etc. Among them all it is possible, if
one will take the pains, to obtain a bouquet of different
colored leaves vying in beauty with those of autumn.

The change of color, especially to yellow, usually begins
at the peripher}^ of the leaf and then follows the chlorophyll
as it recedes, and its nitrogenous elements are absorbed into
other parts of the plant. But this manner is not uniform ;
maples often show the red color first in the veins, and some
while green, and others when changed to yellow, are not
infrequently blotched and variously splashed with blood-red.

Circidiphyllum frequently ])egins to develop its pure
yellow at the base of the leaf. Leaves suffering from in-
sects are apt to change color first at the affected part ; those
developing scarlet, crimson, purple brown, etc., often have
those hues confined to the parts only which are reached by
the direct rays of the sun, while the shaded })ortion of the
leaf remains green or yellow, and the outline of the upper
or shading leaf is depicted as a sun picture upon the lower
one. Examples of the latter are common in Maples, Ampe-
lopsis Veitchei, Gaultheria (boxberry), Vaccinium corym-
bosum (blueberrj' ), Lonicera, etc., and are especially notice-

23

able in the fall. No sooner has the new foliage of the year
assumed its emerald hue than it begins to show signs of
decay in individual cases, the leaves here and there turning-
yellow, orange, russet, red, dull purple or purple brown,
and falling off. Yellow leaves have been seen on maples
even in May.

Various causes operate to produce the change of color ;
the following may be designated : lack of sun and air (ven-
tilation), bad drainage, drought, frost, excessive heat,
injury to roots, branch or leaf, fungus, blight or other dis-
ease, transplanting, rapid evaporation* by drying winds or
hot sun, embrace of twining plants, injuries from insects,
ripening of the leaf, sudden changes of temperature, forest
fires, injuries by storms and other accidents, etc. We shall
illustrate some of the foregoing causes of color change from
actual observations :

A potted India Rubber plant was placed into a lal'ger,
but closely fitting, decorated pot having no outlet for drain-
age. In a short time the leaves began to turn a rich yellow
and fall. It was found that there was an accumulation of
water in the outer pot from excessive watering, wiiich
drained from the inner one, and this had driven out the air
necessary for the roots, for the plant speedily recovered
with proper drainage.

Trees, shrubs and herbs with dense foliage are apt to
have some of their interior leaves ''damp off" and turn
yellow from lack of sufficient sun and air. This is frequently
seen in sugar maples, apple trees, currant and gooseberry
bushes, and among low plants in shady woods even in
early suumier.

Plants in greenhouses also decay from too much watering
during the resting period. The leaves of many gray birches
{Betula popuJifolia, Ait.) growing in dry, sandy ground
were largely yellow early in July ; the trees were evidently
suffering from drought as the month was very hot and dry,
A number of herbaceous plants growing in wet places

24

change color, or at least the lower leaves, when the water
dries up in early summer, as Polygonum, Ludwigia, and
Statice, which change to red or purple.

Also in July a considerable number of leaves of Pyrus
arbutifolia, L. in similar dry soil were seen of a bright red
color ; it was found that nine-tenths of these had been
punctured or otherwise injured by small insects which had
sucked their juices. The latter part of July, Ampelopsis
qninquefolia, Michx. often displays more or less red leaves.
They are most commonly seen in sunny places, especially
on stone walls and dedges. Their color is also induced
partly l)y insects which prefer sunshine to shade, and in
part to excessive heat, direct and reflected, increasing trans-
piration in excess of absorption. The temperature in such
situations is frequently 15 to 20 degrees higher than in the
shade and often above 100 degrees Fahrenheit. A portion
of the same vine if shaded from the sun's rays is likely to
be green and not attacked by insects.

July 20, many yellow leaves were seen on black walnut
(Juglans nigra), and butternut {J. cinerea). They were
caused by a caterpillar | in. long covered with a fluffy,
white, woolly substance which stood erect on their backs.
This was doubtless for protection from enemies, as it finally
disappeared. They bit a small crescent shaped hole about
an inch from the base of the leaf, severing the midrib at
that point and apparently sucked the juices Of the nu-
merous fallen leaves every one had the midrib punctured at
the same spot.

About the m,iddle of August, bright crimson leaves were
seen on a scarlet oak ; this was two months before the leaves
of this oak change color usually. It was found that every
colored leaf had been punctured on the midrib precisely as
those of Juglans I

Some leaves of potted Pelargoniums in February were
noticed to be red at the tips. The broad, heart shaped
leaves, of a succulent character, were found to have been

25

accidentally broken across the brittle leaf ribs \n handling,
or by brushing j:»ast them as they stood near {he sides of
the benches ; only the outer portion of the leaf beyond the
break was changed in color. It was perfectly evident that
the cutting off of the supply of water by the fracture of the
ribs, while at the same time evaporation continued, had
induced chemical changes in the tips similar to those of a
dying leaf.

A fine specimen of Cladrastis tinctoria, Itaf. July 1 ,
bore yellow leaves on one side of the tree and green on the
other. Tussock moths were abundant, but they did not
seem to account for the division of color ; the real cause was
that borers had attacked the trunk close to the ground and
girdled three-fourths of the circumference. By the middle
of August half the limbs, viz., those which bore the yellow
leaves, were bare and the leaves of the other half had
become yellow.

Sept. 1, a Fringe tree (Chionanthus) and a flowering-
dogwood (Cornus) were changing color; the former to
yellow, and the latter to red. Their trunks were riddled
by borers ; near the same time a line Tupelo (Nyssa) was
aglow with crimson and purple brown ; in this instance
borers had attacked the branches and eaten off much of the
bark on the principal limbs.

A red maple when seen, had changed to a yellow hue,
excepting one branch which was crimson and in strong con-
trast with the bright yellow. It was found that at the
bottom where it formed a fork with another limb it was
deformed by a knot much larger than the stem. This de-
formity, probably originally caused by an insect, impeded
the circulation of sap with the result of a change in color.
The singular appearance may have occurred in a previous
season, and would be likely to be repeated in the future.
A Circidiphyllum of a prevailing yellow color, though red-
dish at the top, had a single lower branch of a crimson
})urple I Two months before, the gardener at the sugges-

26

tion of the writer made some incisions on tlie bark, cutting
and partly stripping it down an inch or two ; the result as
seen, was a branch of a red purple instead of yellow. It
would seem that a lack of nourishment or water may pro-
duce a darker color ; if so this may explain why some red
maples of the same age, and growing in the same vicinity,
change to yellow, while others change to scarlet or crimson.

In the middle of August several red maples seen in dif-
ferent localities were ablaze with crimson color. Each tree
was visited and the cause of the color investigated. One
was found to be deeply girdled by the woodman's axe : one
had the bark entirely peeled off to the height of six feet,
probably b}' hoodlums ; another, a beautiful shade tree in
the city, had one magnificent branch of crimson-scarlet,
while the others retained their ordinary verdure. The
owner of the residence before which the tree stood admired
its beauty, but was unable to account for it. The explana-
tion was simple : a horse which had daily stopped at the
tree, had gnawed the bark badly on the color side ! The
leaves of elm and ash in city streets, frequently mutilated
in a similar manner, turn prematurely yellow, and fall.
Such, besides the unsightly appearance of the trunks, are
the effects of utilizing trees in cities for hitching posts.

Several maple trees with bright crimson leaves were
noticed in a wooded swamp. When the^brakes and small
bushes were brushed aside, the trunks close to the ground
were noticed to be somewhat darker than usual, but it did
not seem probable that it could be the result of a fire, as the
soil was still wet, but such proved to be the fact, as a farmer
residing near by stated that one had been set by bicyclists
in February. Two other instances of premature coloring
by forest fires were observed later in dry woods.

Of the numerous kinds of insects directly attacking the
leaves, those most effective as a rule in modifying the color,
appear to be the diminutive ones which pierce minute holes
and very leisurely suck the juices, such as the aphides,

27

rather than those more ravenous like the Elm Beetle and
Tussock Moths, which devour much of the substance of the
leaf, and leave the remainder to dry up and turn a dead
brown.

The early change of Pyrus arbutifolia, Sassafras, Maple,
Birch, Elm, Rhus, Tupelo and Ampelopsis, is largely due
to their agency, and in dry seasons, in dry or sandy soil,
these trees and shrubs are conspicuous for their bright
colors before the light frosts appear in the latter part of
September, but the change is not general, as the gray birch
and elm are really among the last to lose all their leaves,
and even many individuals of maple and tupelo retain their
foliage until the October frosts.

Oak tree borers, after channeling the stem of a small
terminal branch for several inches, make a neat transverse
cut entirely across the ligneous portion, sparing only the
bark, by which sprays two or three feet in length hang sus-
pended with their unsightly, dried up, brown leaves until
dislodged by wind or rain. These dead branches are most
commonly seen on the scarlet or black oak, but are also to
be met with on other species, and on the elm.

Trees transplanted in summer are [)retty sure to have
some of their leaves change color. As an experiment some
young growths of birches, pyrus arbutifolia and rhus copa-
lina were transplanted. The lower leaves of the birch
turned yellow, half of those of rhus and nearly every one of
pyrus became crimson. Disturbance and injury to the roots
was the cause. Colored leaves can easily be produced arti-
ficially by girdling and other methods. Many leaves when
changing color become red or yellow according to exposure,
as the tupelo, which is scarlet in the sun, yellow in the
shade, and orange in medium illumination.

Several red tips of branches of Leucotho'e raceniosa, Gray,
Aug. 10, were due to injury : one branch was broken,
another split down the stem, one eaten by a worm under
the bark, and one was closely entwined by the tendrils of

28

Smilax glauca! Gaylussacia resino.sa is frequently seen
having a stunted branchlet of a rose color from the effect
of a fungus,

Avnjjelopsis Veitchii, which is generally green in August,
was seen in many instances to have long, running terminal
branches of its closely clinging vine with leaves of a bril-
liant red color. These were traced back to a point where
the stem was much constricted for a quarter of an inch or
so and swollen on either side ; as the constrictions ol)served
in different vines were always similar, they were probably
due to insects. Instances were observed where there was a
succession of red leaves from a single root streaming and
intertwining to the highest point among perfectly green
leaves, borne on stems from other roots. The red-leaved
stems were constricted close at the ground. The connecting
link of the constriction was often reduced to a slender
thread which sometimes becomes desiccated, when the
leaves beyond the point fall off ; on the other hand, the
injury in some cases is healed hy enlargement and union of
the parts before the link is broken. A vine may become
red-leaved from other causes. A brick building on its
extensive northern wall was mantled with green Ampelopsis
variegated with much red color. The red leaves originated
from a stem whose roots probably penetrated into gravelly
or stony ground where there was a deficiency of moisture.

The southern side of a church clad with Ampelopsis
which was exposed to the full glare of the sun, became a
solid red color, also late in August. Undoubtedly this
early change was also caused by drought. Another exam-
ple of Ampelopsis coloring was peculiar in having an inter-
mediate portion of a running branch with small, red leaves,
while both below and above on the same stem the leaves
were green and of a larger size ! A probable cause of this
phenomenon was, that when the now red leaves were green
and terminal, they lacked water and became dwarfed and
burned by the sun, but with more favorable conditions, the

29

later leaves attained their normal [)r()portions. A similar
irregular and interrupted growth is not infrequently seen in
ferns when several of the central pinn.e of the frond are
much reduced in size, and for similar reasons.

In spring a vine of Ampelopsis on the north wall of a
building was wholly green, but where it turned a corner to
the east side it was a purplish brow^n. A simiUir change of
color was witnessed in the fall on a vine on a brick church.
It was scarlet on the south side wiiich had direct sunlight,
but green on the west. Even a change in the declination of
the sun under certain conditions may produce a change of
color. A vine on a north wall of a building had become
scarlet from drought and three or four hours of afternoon
sunlight. When the sun no longer shone upon that side,
the later growth of runners bore only green leaves ! More-
over on a building wall composed of boulders, or cobble
stones, the leaves on each rounded surface were colored,
but those in the depressions between, which were of mortar
and partly shaded were green. The effects of frost on Am-
pelopsis will be referred to farther on.

Ampelopsis Veitchii is a very interesting plant for the
study of the effect of insects, drought, frost, etc., on the
color of leaves.

Allusion has been made to the fact that there is a natural
coloration of ^s foliage common to certain species in early
spring, but dependent upon the amount of direct sunlight
which they receive.

The foliage of the cultivated, red variety of Begonia
vernon growing in the shade is green, but in the sun is a
dark mahogan}^ color on the parts of the leaf reached by
the direct rays, as a side, the tip, along the midrib, or on
the lower surface, according to the i)osition of the leaf,
which has its sides naturally more or less incurved, thus
preventing the impact of the sun's rays upon the entire
surface. Varieties of purple-leaved beets and lettuce, of
Ampelopsis Veitchii, Canna. etc., are similarly dependent

30

for color: and the purple-leaved beeches, birches, and bar-
berry, etc., have a deeper color in the sun. The stems,
peduncles, pedicels and calyx of Phytolacca decandra, L.
turn from green to a deep crimson. The coloration of
many other stems and petioles on their sunny side, and the
rosy cheek of the apple, pear, and peach is of a similar
nature. These, as also the spring examples already men-
tioned, are the colors of health and vigor, and should be
distinguished from the tints of the dying leaf ; they are at
times however so similar that it is difficult to determine to
which class they belong. The development of color in
flowers themselves is not unlike that of the parts of the
plant mentioned. In Begonia vernon the red color of the
flower was seen in the foliage, only that it was modified by
the chlorophyll. The white and pink varieties of this plant
have their foliage unchanged by sunlight.

Florists are often able to determine the color of the future
flower by the tint or shade of the leaf.

Coleus, Acharanthus and similar foliage j^lants are so
surcharged with their coloring ingredients as not to be
dependent upon strong sunlight, but the greater part of
evergreens, which become more or less changed, and the
brightest hues of autumn require direct sunlight as a factor
in connection with other inducing agents.

The golden leaved Elderberry changes its youngest leaves
from green to yellow. Yellow with few exceptions does
not require the sunlight to develop, but it is of a richer
color with it.

The season for autunmal tints may, for convenience, be
divided into three periods : from the middle of August to
the middle of September, from the middle of September or
the first lio-ht frosts, until the middle of October, and the
last period extending to middle of November, or for ever-
greens, into December. In the first, it is virtually summer
protracted into September ; usually a very dry season with
little or no rain, when vegetation often suffers from drought.

31

One may then expect the red maples and elms to begin to
show color on dry uplands, also in similar places the choke-
berry (Pyrus arbutifolia), tupelo (Nyssa sylvatica), Vir-
ginian creeper (Ampelopsis), and white birch (Betula poi)-
ulifolia, Ait.), etc.

Early in September, before there had been any frost, a
southern hillside was brilliant with the crimson foliage of
the swamp or red maple. The thermometer rarely registers
so low as 32° Fahrenheit in Massachusetts in the first part
of this month, but it does occasionally in the latter part.
The cause of this premature coloration was obviously the
prevailing drought, affecting especially uplands, for it was
remarked that an extensive row of the same species in a
well watered meadow at the base was perfectly green.

It was the hot weather and drought, together with the
injury sustained from insects, which had also induced the
colors in the other species above mentioned, for the same
kinds under more favorable conditions as to moisture con-
tinued green. That this was a correct view of the matter
was proved by numerous observations in different localities ;
the trees and shruljs with colored leaves being always in
dry situations.

In the second period the weather conditions are consider-
ably different: the fall rains set in, the nights become cool
or even chilly, and a hoar frost is likely to occur in the
latter part of September. There was a light frost Sept. 20
of the present year 1900, the thermometer registering in
different localities from 36 to 39'' Fahrenheit. The effects
of this frost would hardly have been noticed except by a
close observer, for the reason above stated, that several
kinds of trees and many shrubs had already assumed autum-
nal hues ; some instances however were noted which served
to show that the foliage had been subjected to some other
influence than drought : Hydrangea Japonica, var. Hortensia
which till then was green, became much bronzed and later
changed to red ; the yellow leaves on elms, ashes and

32

birches became more numerous, also the foliage of many
wild shrubs became conspicuous for brighter hues ; an
ampelopsis vine on a lofty building became crimson in the
course of a few days ; a broad belt on the lower side of a
field of millet changed to a pale, grayish green, while the
upper portion, separated almost by a straight line, retained
its normal deep green color ; but a matter of more conse-
quence from an agricultural point of view was, that a farmer
not many miles away lost one half of the cranberries in his
bog ! It is during this second period that the red maples in
valleys and lowlands assume their most brilliant hues, cul-
minating in splendor from the 10th to the 15th of October.
Other plants conspicuous for color at this time are cornus
florida, witch hazel, yellow birch, several species of rhus
(poison dogwood and poison ivy), sassafras, clethra, grape
vine, smilax, hickory, tupelo, blackberry, blueberry, choke-
berry, hazelnut, Virginian creeper, etc. The landscape in
south-eastern Massachusetts during this period for the year
1900 was especially gorgeous with a profusion of the most
lovely tints. Upon the appearance of frost in October we
enter upon the final period for deciduous foliage. In the
present year on the night of

Oct. 16, the thermometer registered 34° ; ice was seen in one locality.
" 17, "■ '• " 34°; there was a hoar frost.

" 19, •• "• " 28° ; ice and frost were formed.

" 20, " " '• 34°; frost was noticed.

These were the minima temperatures shown by a self-
registering thermometer. The temperature varies greatl}^
in different localities. The above were registered in the
city proper: in the country, away from the seashore, it is
often lower.

While the landscape just before these frosts was resplen-
dent with gay colors induced by the double agencies of
drought and frost, a little attenticm to the subject disclosed
the fact that perhaps not much more than fifty per cent, of

33

our trees, shrubs, etc., had changed color up to that time,
especially so if one includes the numerous cultivated species.

The following is a partial list of plants continuing green
until the frosts of Oct. 16-20 : Oaks, nine species, poplars,
three native and three cultivated, willows, many species,
black walnut, butternut, Ijuttonwood (Platanus occidentalis
and P. orientalis), beech, chestnut, locust (liobinia Pseud-
acacia), honey locust (Gleditschia), wild cherry (P. sero-
tina), English elm, slippery elm, Dutch elm, English oak,
linden, horse chestnut, white birch mostly green, mulberry,
beach plum, elderberry, privet, osage orange, buckthorn,
white thorn, Norway maple. Sycamore maple, white maple,
black alder (Alnus), winterberry (Ilex), magnolia, catalpa,
sophora, etc. : many garden shrubs : althiva, Cornelian
cherry, laburnum, wistaria, Forsythia, pyrus Jai)onica,
lonicera, hydrangea paniculata, clematis, deutzia, spiraea,
ampelopsis Veitchii (the most of these vines green), phila-
delphus, bittersweet (Celastrus), weigelia, bignonia ; fruit
trees : apple, pear, cherry, etc.

Some of this list scarcely change color, as mulberry and
catalpa, others being semi-evergreens, change only to dull
or sober hues.

Immediately after the frosts of October the foliage of the
trees which had previously changed color became less bril-
liant, the yellows being somewhat browned, the reds dulled
and many trees defoliated. The remaining foliage was
nevertheless still beautiful, though more quiet and subdued.

The succeeding frosty nights, which were followed by
warm days, were, however, rapidly bringing new beauties
into evidence. Special observations were made just before
and after the frosts on certain prominent, vigorous trees
and vines which had retained their verdure until then :

The green leaves of the terminal runners of Ampelopsis
Veitchii on the north side of the church before mentioned,
were found two days later to have changed to red ; at the
same time the green clad western side of another church

34

became completely crimson ; certain red maples which being
favorably situated had retained their emerald hue, glowed
with crimson and gold ; a cluster of shrubs of Rhus typhina,
L., which previously had only a few red, lower leaves,
became entirely crimson ; a scarlet oak, being near the
writer's residence, had been carefully watched from day to
day. It had no red leaves at all before the frosts, l)ut, two
or three days after, all of the foliage of the outer branchlets
were tinged with crimson, and in less than a week every
leaf had changed color. The leaves of a white oak in the
vicinity changed to a yellow russet above and purple be-
neath. Four handsome Tupelos of large dimensions, spared
when the woods were felled and which the writer passed
daily, changed almost in a night to deep crimson. Sugar
maples, not native here but planted extensively for shade
trees, assumed a mellow yellow, also elms, which had been
slowly shedding their leaves since August, now changed
entirely to yellow ; Liriodendron, Honey Locust, Kentucky
Coffee Tree and Horse Chestnut joined the gay assemblage.
Thus "millions of leaves were painted by the magic fingers
of Frost with innumerable and indescribably beautiful tints
and shades of color " in the course of three or four days !

Extensive forests principally of oak, and almost entirely
green before, also when viewed a few days later from a
commanding eminence, except for a few pines, were com-
pletely changed to a reddish brown or brick red color ! No
one who had viewed these trees before and after the frosts
could have had any doubts as to the cause of the wonderful
transformation. And yet it is asserted with much positive-
ness by some writers that frost has no effect upon the color
of foliage !

J. Y. Bergen, in Elements of Botany, states that "The
brilliant coloration, yellow, scarlet, deep red and purple of
autumn leaves is popularly but wrongly supposed to be due
to the action of frost.'" This positive statement is in a few
lines farther on (|ualiiied by the remark that " Frost perhaps

35

hastens the breaking up of the chlorophyll, but individual
trees often show bright colors long before the first frost,'" etc.

He, it appears, is not certain that frost has any effect
upon color, and seemingly can assign no reason for the
change of color before frost. Mr, Bergen's ideas and lan-
guage are very similar to those of Doct. George L. Goodale,
a much earlier writer, in his PJiyHological Botany, who
says: "That frost is not essential to the production of the
leaf-colors of autumn is plain from the widely known fact,
that many leaves undergo precisely these changes of color
lo7ig bejore any frosts appear. It is generally believed how-
ever that f-eezing may somewhat hasten the process of chloro-
phyll disintegration which underlies all the changes."

Geo. B. Emerson in Report on the Trees and Shrubs of
Massachusetts, writes as follows regarding the Scarlet Oak,
p. 166, vol. 1 : "The leaves after they have undergone this
change of color vjhich has no dependence on the action of frost,
remain long upon the tree," etc.

The Scarlet Oak here, in the vicinity of New Bedford,
was scqj'cely changed at all before the frosts of Oct. 16-20.
A letter from the Arnold Arboretum, dated Oct. 23, 1900,
states that: "The foliage of the Scarlet Oaks is just be-
ginning to turn here." Advices from other points in New
England were of a similar tenor.

Again quoting Emerson, p. 552, vol. 2: "The observa-
tions for a single year of the varying colors of the red maple
Avould be sufficient to disprove the common theory that the
colors of the leaves in autumn are dependent on the frosts.
It is not an uncommon thing to see a single tree in a forest
of maples turning to a crimson or scarlet in July or August
while all the other trees remain green. A single brilliantly
colored branch shows itself on a verdant tree, or a few
scattered leaves exhibit the tints of October while all the
rest of the tree and wood have the soft greens of June,"
etc. "The frost has very little to do with the autumn
colors." But, the "single tree," "single brilliantly colored

36

branch," "or a few scattered leaves" in August mentioned
by Emerson have been fully accounted for in the preceding
pages, and it is unnecessary to remark further that such
cases have nothing whatever to do with the agent that trans-
forms the color of a whole forest almost in a night !

Of course it is to be understood that it is not frost alone,
but the alternate action of frost and sun, the sudden changes
of temperature, which derange the normal functions of the
leaf, and induce the chemical changes which accompany the
change of tint.

Gray's Manual states that Quercus rubra, L. Red Oak,
turns dark red after frost !

A hoar frost, as is well known, "is the moisture of the air
condensed at freezing temperature upon plants and other
objects near the surface of the earth." It is not limited to
a general temperature of 32° Fahrenheit or below, but " maj^
as a rule be expected when the thermometer indicates S'^ to
10" above the freezing point"! A. G. McAdie states it is
liable to occur at an air temperature of from 40° to 45° Fah-
renheit when the other conditions are favorable, as a clear
and still night and much moisture in the air.

The cause of the frost is the radiation of heat which re-
duces the temperature of the surface of the earth and vege-
tation below that of the free air. Rapid evaporation also
reduces the temperature, as is shown by the difference often
seen between the wet and dry bulb of thermometers, which,
at times, may be as much as 12°, or more, lower in the wet
one, which more nearly represents the actual temperature
of foliage wet with night dews.

The minimum temperature at New Bedford, Mass., on
three different nights for September, 1900, was SG-^, 39°,
43°. Upon an examination of the records of the City En-
gineer's Department for the preceding nine years, it was
found that in each year there were at least three days when
the minimum temperature was not above 42°, and ranging
from 35° to 42°, except in 1891, when the minima were 44°,

37

46o, 49°. The reports from various stations in the State to
the Weather Bureau at Boston for Septembers average con-
siderably lower than this, thus : in 1896 from reports of 27
stations there were 10 under 32*^, 25 under 40°, the lowest
being 30°. In 1897 of 24 stations, there were 2 under 32°,
21 under 40"", the lowest being 31. With the same ratio
for the entire State, it appears therefore that there would be
freezing weather in a considerable number of towns every
September, and that light frosts would be liable to occur in
almost every place !

The difference in the mean temperature of the summer
months is generally insigniticant. In 1895 there was an
increase of 2° l)etween that of June and of July, and 2° more
between July and August. In 1898 the increase between
the mean of June and of July was 6° plus, and between
July and August but eight-tenths of a degree. Between
August and September, however, there is usually a decrease
of 8°, and between September and October 12°, the mean
tem[)erature of October being on an average 20"- below that
of August !

Growers of roses find a temperature of not lower than
55° at night and 70"^ to 75° by day the best for certain va-
rieties, as Brides and Bridesmaids. If the temperature
should fall 10° lower at night, the growth would be checked,
few or no roses produced, and the plants be liable to mildew
and black s[)ot and to change color. That is to say, the
sudden change from an unusually low temperature at night
to a high one would be prejudicial to the health of the plant.

If a plant under glass is injuriously affected by a fall of
10° below the usual temperature, it is quite likely that an
average fall of 20° in the open air would diminish the vital-
ity of many kinds of leaves, and that the sudden change
from very cool nights to very warm days would be likely
to slowly disintegrate the chlorophyll and induce the autum-
nal tints even when there were no frosts I Notes on Frost,
by Prof. E. B. Garriott, Weather Bureau, Washington,

38

says: "Cotton will be seriously injured by a low tempera-
ture early in the spring whether frost occurs or not." "No
adequate means of protection against cold and frost, suitable
for general use, has been discovered." Another writer
(John Tyndall) says : "Sudden changes of temperature are
prejudicial to animal and vegetable health." This point
may be farther illustrated by the following instance : A
florist early in May discontinued the fire in his hot house
devoted to Pelargoniums, as he had but a small supply of
fuel on hand. His method was to close the ventilators early
in the afternoon and thus store up heat for the night. Up
to the time that the fire went out the plants Avere healthy
and green, with only an occasional yellow or reddish leaf.
In about a week after he ceased firing, all the lower leaves,
being fully one-third of the total foliage of the central bed
and a part of the south side, changed to red ! Those on the
north side, and those close to the south side which were
partially shaded by the framework of the roof and side,
retained their original verdure. The central beds had the
full benefit of the direct rays of the sun. The day tempera-
ture was probably a little higher in the afternoon, and a
little lower in the morning, but averaging about the same
as before for the daytime, while the night temperature was
probably 10"^ or 15° lower, as there were some frosty nights.
The plants were ventilated and watered according to require-
ments, and it is believed that there could not have been any
frost within. That the red color was induced by the changes
in temperature there could be little doubt. They had pre-
viously experienced as much heat by day, and the only spe-
cial difference was the lower temperature at night. It has
been mentioned that the plants partially shaded were unaf-
fected; it was only those subjected to hoth extremes of heal
and cold that became red. As a further test of the real
cause of the red leaves, several of the potted Pelargoniums
and also of Ampelopsis Veitchii were plunged into the
ground in the open air. There was a hoar frost soon after

39

and in less than a week the Ampelopsis became purple and
the Pelargonium red ! No red leaves appear on these plants
during the summer months, but an occasional leaf dying
from any cause simply turns yellow. The rationale of the
phenomenon would seem to be, that all the jilants being
somewhat chilled by the low temperature and dampness, the
older leaves of the central plants receiving the full power of
the morning sun, on account of their diminished vigor and
the check to the flow of sap, were unable to respond to the
demands of an active evaporation caused by a hot sun.
"The drying of the tissues is fatal to the component cells
And the organic contents speedily undergo decomposition,"
and 1 may be permitted to add to the quotation : that new
colors result from the chemical changes set up. This 1
apprehend to be the explanation of the color change hi the
Ampelopsis and Pelargonium, and also of the effect of frost
as well as that of drought, and it will apply with equal cor-
rectness to some at least of the winter colors of evergreens
which will now be considered. One of our cultivated vines
which retains its leaves among the latest is the semi-ever-
green Lonicera Halleana or Japanese Honeysuckle. It may
be seen at times as late as Christmas, more or less variegated
with dull reds, reddish brown or yellowish colors, together
with much of the unchanged green foliage, presenting a
cheerful sight while most trees and shrubs are leafless. But
not every Lonicera is thus changed in color, for many a
trellis covered with this vine bears only green leaves. It
depends upon the exposure. A trellis serving the purpose
of a fence or screen covered on both sides with vines was
colored only on the sunny side, the other, more exposed to
the cold north winds, remained entirely green ; also similar
vines when on the north side of dw^ellings were found to be
green and when on the south they were colored !

Many of our native shrubs of an evergreen character are
similarly affected in the winter months: GauUheviu procum-
hens, Smilax glauca, Kalmia auguxtifolia and Ilex glabra on

40

the sunny side of wooded roads are apt to be colored various
hues as red, purple brown, or jet black, while similar plants
on the opposite and shaded side retain the normal green
color.

If a leaf of Lonicera chances to be turned so that the
inner side is outermost and faces the sun, it will assume a
deep blue color, much different from the usual purple-brown
of the other side, this being on account of the different
character of that surface. If it happens that one leaf par-
tially shades the one below or even at some distance in the
rear, the shaded portion will simply change in the usual
manner, commonly to yellow, while the projecting part,
whether tip, side, or entire border, receiving a more intense
heat will, in the case of a -maple, be likely to become scarlet,
or in Lonicera a purple brown, and in Ilex an inky black ;
the contour of the upper leaf, whether rounded or angular,
will be photographed as before stated upon the lower one,
and when from intense cold the sides of Lonicera become so
strongly reflexed as to become tubular, the sun paints only
a longitudinal streak along the prominent rounded part.
All of these evergreens are subjected to substantially the
same degree of cold, but cold alone, at least as a rule, is
incompetent to change the green color; it is the freezing
and rapid thawing in the sun of the leaves, causing evapo-
ration faster than absorption, which causes disruption of the
chlorophyll.

Other evergreens which [)artially change color from these
causes are Mahonia, Forsythia, Rumex, Sphagnum, Black-
berry, Cranberry, Pyxadanthera, Pipsissewa, Sheep Laurel,
Privet, Strawberry, Pyrola.

According to Kraus, the chlorophyll of certain evergreens
is not disintegrated by sun and frost, but a reddish or
yellowish substance develops, supposed to be tannin, which
conceals the unchanged green granules during the cold
season. This may refer to the class of evergreens includ-

41

ing Arbor Vitoe, Retinispora, Red Cedar, and others assum-
ing a yellowish or brownish color.

Hemlocks and spruces generally retain their verdure in a
remarkable degree. The variegated Japanese Honeysuckle
has its recent leaves of a yellow color, or green reticulated
with yellow. It is only after successive frofit!< in Noveml)er
that the yellow is replaced by crimson !

Euphorbia Cf/jjarissias, a garden escape, often common
by roadsides, also changes only after freezing weather in
November, from green, through purple, to a bright orange
red !

"Chlorophyll during the absorption of light is slowly
broken down. If it is decomposed faster than it can be
rebuilt by the protoplasm the entire leaf dies." (Prof. I).
T. Macdougal).

"Another causie which may disturb the relation between
absorption and transpiration, is the diminished conductivity
of woody tissue at low temperature." (Doct. G. L. Good-
ale).

The autunm change of color in foliage, resulting in part
from the decomposition of the chlorophyll, in distinction
from the screening colors which disappear after a time, indi-
cates a loss of vitality and a gradual dying of the leaf.
This state can evidently be effected by either drought or
frost, as well as by many other causes. In the case of
drought, the lack of moisture at the roots prevents the
necessar}^ supply of water to the leaves, on account of which
the tissues become dried, and the character of the cell con-
tents greatly changed. In the case of a frost, the frozen
ground, or chilled trunks and twigs, with lessened conduc-
tivity, fail to deliver to the leaves an adequate amount of
water for transpiration, and the result to the leaf is virtually
the same as from drought.

If the various causes of color-change enumerated in the
preceding pages were analyzed, it would be found that the
proximate cause in all cases was an insufficient supply of

42

water to the leaf. Frost alone, as previously remarked,
may not be competent in all cases to effect this change, as
in the semi-evergreens, Lonicera Halleana, etc., the reason
being probably that when frozen roots, stems and leaves are
in the shade and the atmosphere be not too dry, there would
be little or no transpiration, and with return of milder
weather the leaves might wholly regain their lost functions
without any change of color or serious damage to their
tissues.

It is undoubtedly the frosty nights followed by warm suns
that changes the foliage of the oaks and all other remaining
foliage capable of change, represented by a large number of
different species. Before the frosts arrive, individual oaks
may change from drought the same as other trees, but whole
forests do not often so suffer, and yet they regularlj^ change
color rapidly and within a few days after the severe and
late frosts of October. About November 5 they are iii their
brightest and best tints of reddish brown, yellowish brown,
purplish tan, dull purple gray, and similar tones.

December 1, 1900.

43

CONTENTS.

rO "OBSERVATIONS OX THE COLORS OF LEAVES.

Replies to CiiticisiiLs,

Color in Animals, ....

Redness of Skin from friction not due to

ulation, .....
Nerves and Irrital)ility, .
Transferability of Color,
Natural and Artificial Change of Color,
Observations on the Colors of Leaves,
Autumnal Colors before Frost,
Autumnal Colors after Frost,

stnn-

3

7

8
11
15
16
21
30
33