Morren[233] describes a form intermediate between the ordinary slipper-shaped corolla and the perfect peloria just described, and which he calls sigmoid peloria. This flower is intermediate in direction between the erect peloria and the ordinary reflected flower. The tube is curved like a swan's neck and is dilated in front into two hollow bosses, such as we see in the lower lip of an ordinary flower; beyond these it is contracted and is prolonged into a slender beak terminating in two hollow teeth, between which is the narrow orifice of the corolla. The colour at the base of the tube inside is as in the perfect peloria; while round the summit of the tube, in both cases, the intensity of colour is greatest on the outside. Now, in a normal flower the deepest colour is within just opposite the orifice of the corolla; this deep colour is also seen outside of the central and most elevated portions of the lower lip. In the peloria the deep colour at the base of the tube represents that which is near the orifice under ordinary circumstances, while the outer patch of colour at the apex corresponds to that formed on the upper surface of the lower lip. On the other hand, in peloric flowers of Cytisus Laburnum, Clitoria Ternatea, Trifolium repens, and other Papilionaceæ, it is the "standard," the form of which is repeated. In the case of peloric aconites[234] the lateral and sometimes the inferior coloured sepals assume the hooded form usually peculiar to the upper sepal only, the number of the petals or nectaries being correspondingly increased. Balsams become peloric by the augmentation in the number of spurs.[235] So when orchids are affected with irregular peloria it is the form of the labellum that is repeated, the accessory lips being sometimes the representatives of stamens, which are usually suppressed in these flowers,[236] but at other times the appearance is due simply to the fact that all three petals assume the form usually confined to the lip, the staminal column being unaffected, except that its direction and relative position with reference to the other parts of the flower is different from ordinary. This was the case in some flowers of Phalænopsis equestris sent to me by Mr. Wentworth Buller. Fig. 123 represents a flower of Aristolochia caudata with two lips, for which I am indebted to Mr. W. H. Baxter.

From these cases it is evident that the flowers in question become regular by the repetition of the irregular parts.

It is probable that peloria may occur in any habitually irregular flower, and that, if more attention were directed to the subject, illustrations might be obtained from a larger number of natural families than can be done at present. It is, however, necessary to exercise discrimination, and not to attribute to peloria all the cases that at first sight appear to be so referable. Thus, Professor Dickson exhibited at the Botanical Society of Edinburgh, December 13th, 1860, four abnormal flowers of the common Indian cress (Tropæolum majus), each presenting a supernumerary spur. On these he remarked that "in Tropæolum the posterior part of the receptacle between the insertion of the petals and that of the stamens is dilated so as to form the spur which is so characteristic in the genus. The position of the spur in a line with the posterior sepal has led many botanists to consider it as a process of that sepal, but the fact of its being situated within the insertion of the petals is conclusive as to its receptacular origin. In the flowers exhibited the supernumerary spur (as if to show its want of connection with any sepal) was placed exactly between a lateral sepal and one of the anterior sepals, sometimes on the one side of the flower and sometimes on the other. These additional spurs were precisely similar to the normal ones, except that they were a little shorter. This abnormality, although at first sight seeming to indicate a pelorian tendency, is no approximation to regularity, from the fact of the extra spur being differently placed, with regard to the sepals, from the normal one."

Peloria of this kind, when perfect, is very often associated with other alterations. Change of direction is one of the most common of these; the usually drooping flower becomes erect, the stamens and style also are changed in direction, while, not unfrequently, either the one or the other (most often the stamens) are entirely suppressed. With this suppression an increase in the size of the flower very generally coincides. The number of parts is also frequently increased; thus, in Antirrhinum majus the corolla, when subjected to peloria, is very generally six-parted, and has six stamens. Fusion of one or more flowers is also a common accompaniment of peloria, as in Digitalis purpurea, in which plant prolification often adds increased complexity to the flower.

It has been stated by Moquin and others that the uppermost flower of an inflorescence is the most subject to peloria; the uppermost flower of Teucrium campanulatum, for instance, is very generally regular. In Calceolaria it is the central terminal flower which is usually peloriated; on the other hand, in Linaria and Antirrhinum the lower flowers, or those on the secondary branches, are quite as often affected as the primary ones. Cassini considered that the spur of Linaria was developed from the lower petal rather than from the upper ones, because there is more room on the side of the flower farthest from the stem than on the opposite side. With reference to this point, M. Godron remarks that in habitually irregular flowers the apex of the peduncle is oblique, and hence the flowers are bent downwards or spread horizontally, but if the receptacle be quite flat and level then the flower is regular. The oblique position causes some of the organs to press on others, and hence induces abortion and suppression of some parts and increased growth in others that are not subjected to pressure. In a terminal peloriated flower of aconite, described by this naturalist, the flower was removed so far from the nearest bracts that all its parts had the chance of growing regularly. In ordinary cases M. Godron considers that the compression of the lateral bracts is the cause of the irregularity of the andrœcium and of the receptacle.[237]

It has also been somewhat too generally stated that peloria occurs principally on luxuriant vigorous plants. It seems quite as often to happen in plants characterised by their deficiencies in this respect. On this point M. de Melicoq[238] says, referring to Linaria vulgaris affected with peloria, that on the weakest plants the peloriated flower was at the top of the stem; while in stronger plants, with more numerous flowers and larger foliage, the peloriated flowers were principally to be found in the centre and at the base of the inflorescence, and their pedicels were much longer than usual.

Linné, as has been already stated, considered these flowers to be sterile, and only capable of multiplication by division of the root, but Willdenow obtained seeds from the Linaria which reproduced the anomaly when sown in rich soil. Baron Melicoq obtained similar results.[239] Mr. Darwin[240] raised sixteen seedling plants of a peloric Antirrhinum, artificially fertilised by its own pollen, all of which were as perfectly peloric as the parent plant. On the other hand, the same observer alludes to the tendency that these peloric plants have to revert to the usual form, as shown by the fact that when the peloric flowers were crossed with pollen from flowers of the ordinary shape, and vice versâ, not one of the seedlings, in either case, bore peloric flowers. Hence, says Mr. Darwin, there is in these flowers "a strong latent tendency to become peloric, and there is also a still greater tendency in all peloric plants to reacquire their normal irregular structure." So that there are two opposed latent tendencies in the same plant. A similar remark has been made with reference to malformations in general by other observers.

It would be very interesting if some competent naturalist would collect information as to whether any variations in degree of fertility exist in the three forms of flowers in Linaria, viz. the ordinary one-spurred form, which is intermediate between the spur-less and the five-spurred form. It must be remembered, however, that in the latter cases the stamens are often deficient. In the Compositæ, where there are regular flowers in the disc and irregular ones in the ray, sexual differences, as is well known, accompany the diversities in form.

To Mr. Darwin the author is indebted for the communication of some flowers of Corydalis tuberosa (figs. 124, 125), provided with two spurs of nearly equal size. To these flowers allusion is made in the work already quoted[241] in the following terms:—"Corydalis tuberosa properly has one of its two nectaries colourless, destitute of nectar, only half the size of the other, and therefore to a certain extent in a rudimentary state; the pistil is curved towards the perfect nectary, and the hood formed of the inner petals slips off the pistil and stamens in one direction alone, so that when a bee sucks the perfect nectary the stigma and stamens are exposed and rubbed against the insect's body. In several closely allied genera, as in Dielytra, there are two perfect nectaries; the pistil is straight, and the hood slips off on either side, according as the bee sucks either nectary." In the flowers of Corydalis, which were provided with two perfect nectaries containing nectar, Mr. Darwin considers that there has been a redevelopment of a partially aborted organ, accompanied by a change in the direction of the pistil, which becomes straight, while the hood formed by the petals slips off in either direction, "so that these flowers have acquired the perfect structure, so well adapted for insect agency, of Dielytra and its allies."

Peloria, then, is especially interesting physiologically as well as morphologically; it is also of value in a systematic point of view, as showing how closely the deviations from the ordinary form of one plant represent the ordinary condition of another; thus, the peloric Calceolarias resemble the flowers of Fabiana, and De Candolle,[242] comparing the peloric flowers of Scrophulariaceæ with those of Solanaceæ, concluded that the former natural order was only an habitual alteration from the type of the latter. Peloric flowers of Papilionaceæ in this way are indistinguishable from those of Rosaceæ. In like manner we may trace an analogy between the normal one-spurred Delphinium and the five-spurred columbine (Aquilegia), an analogy strengthened by such a case as that of the five-spurred flower of Delphinium elatum described by Godron.[243] The Corydalis, before referred to, is another illustration of the same fact, the structure being the same as in Dielytra, &c.

The ordinary irregular flowers may possibly be degenerated descendants of a more completely organized ancestor, and some of the cases of peloria may therefore be instances of reversion; some ancient Linaria may, perhaps, have had all its petals spur-shaped, and the cases of irregular peloria now found may be reversions to that original form. When both regular and irregular forms of peloria occur on the same plant, as they frequently do in Linaria, the one may be perhaps considered as a reversion to a very early condition, the other to a later state, when all the petals were irregularly formed. But before we can assert the truth of this surmise we must have better evidence as to what the original condition really was than we have at present.

The proximate cause of irregular peloria has been considered to be excess of nourishment, but evidence as to this point is very conflicting. Willdenow states that "radices peloriæ, solo sterili plantatæ, degenerant in Linariam," ('Sp. Plant.,' iii, p. 254); but this opinion is counterbalanced by that of others, while the frequent existence of both forms on the same plant, at the same time, seems to negative the supposition of any direct effect from external circumstances.

The following are the plants in which irregular peloria has been most often observed:

The literature of peloria is very extensive. The following are the principal papers, not already mentioned, which relate to the subject, arranged under the genera, placing those first which are most subject to this anomaly (see also Regular Peloria).

PART III.
METAMORPHY.

The term metamorphosis, then, really implies an alteration in the organizing force, taking effect at a very early period of the life of the flower, at or before the period when the primitive aggregation of cells, of which it is at that time composed, becomes separated or "differentiated" into the several parts of the flower. In other words, the "development" of the flower pursues a different course from what is usual. In the preceding sections the effects of arrest and of excess in this process have been partly treated of; other deviations arising from similar causes will be mentioned elsewhere, but, under the present heading, are specially included cases not of merely diminished or increased, but of perverted development; the natural process is here not necessarily checked or enhanced, but it is changed. Hence, in the present work, the term metamorphy is employed to distinguish cases where the ordinary course of development has been perverted or changed. As it is applied solely for teratological purposes, the ordinary acceptation of the term, as nearly synonymous with "development," is not interfered with.

In order to avoid other possible misapprehensions, the terms retrograde and progressive metamorphosis employed by Goethe are not herein used, their place being, to a great extent, supplied by the more intelligible expressions arrest or excess of development.[244]

CHAPTER I.
PHYLLODY.

This condition, wherein true leaves are substituted for some other organs,[245] must be distinguished from Virescence, q. v., in which the parts affected have simply the green colour of leaves, without their form or structure. The appearance of perfect leaves, in place of other organs, is frequently looked on as due to retrograde metamorphosis, or to an arrest of development. But this is not strictly correct; for instance, suppose a petal, which is very generally merely the sheath of a leaf, with the addition of colouring matter, to be replaced by a perfect leaf, one in which all three constituent parts, sheath, stalk, and blade, are present, it surely can hardly be said that there has been any retrogression or arrest of development in the formation of a complete in place of an incomplete organ. The term retrograde here is used in a purely theoretical sense, and cannot be held to imply any actual degradation. Morphologically, as has been stated, the case is one of advance rather than the reverse, and hence the assignment of instances of this nature to a perversion of development, rather than to a diminution or to an exaltation of that process, seems most consistent with truth. The affected organs have really undergone no actual change, simply the direction of the organising force has been altered at a very early state, so that the usual differentiation of parts has not taken place.

Phyllody of the bracts.—As bracts are very generally imperfect organs, so their replacement by perfect leaves is not attributable to arrest of development or retrograde metamorphosis, but the reverse. The bracts of some species of Plantago[246] are very subject to this change. Thus, in the rose plantain of gardens, P. media (fig. 126), the bracts are leafy and the axis depressed or not elongated, so that it is surmounted by a rosette of small leafy organs. A similar condition of the bracts, unattended with arrest of growth in the axis, is common in P. major (fig. 127) and in P. lanceolata (see p. 108). It also occurs in the bracts of Corydalis solida, Amorpha fruticosa, Ajuga reptans, Parthenium inodorum, Centaurea Jacea, in the involucral bracts of the dandelion, the daisy, and many other composites. In the 'Gardeners Chronicle,' 1852, p. 579, is figured a dahlia in which the bracts of the involucre and the scales of the receptacle had all assumed the form, texture, and venation of leaves.[247]

In Umbelliferæ the substitution of leaves for involucral bracts is not infrequent. It has been observed among other plants in Angelica Razoulzii, Carum carui, Daucus Carota, &c. The scales of the hop (Humulus Lupulus) not infrequently manifest this change, as do also the bracts of many amentaceous plants, e.g. in the male catkins of the walnut, the female catkins of the alder,[248] of some willows,[249] &c. The bracts of some Euphorbiaceæ, as E. pusilla, E. Lathyris, E. Cyparissias, have been observed to undergo a similar alteration.[250]

Amongst monocotyledons an analogous change occurs not unfrequently, as in some commelynaceous plants, e.g. Tradescantia, in Musa, &c.

The spathe of Arum maculatum is sometimes represented by a stalked leaf similar to that which occurs, under ordinary circumstances, in Spathiphyllum, but in which genus the spadix is more or less adherent to the leaf-like spathe.[251] In Schœnus cephalotes a similar exaggerated development of the bracts is figured by Rottboell.[252]

Phyllody in inflorescence of Conifers.—This demands passing notice by reason of the interest attaching to the morphological construction of these plants. The elongation of the axis which occurs in the female cones has been already alluded to under the head of prolification of the inflorescence. This change is frequently associated with a more or less foliaceous condition of the bracts, which, indeed, may be seen to be serially continuous, both above and below, with the ordinary leaves. The scales, too, become notched and bipartite, and show, between the lobes, the rudiment of a bud, which in a further stage becomes developed into a shoot bearing leaves. Such a change has been described by Parlatore in Abies Brunoniana, and examples may frequently be met with in the larch (Larix europæa), and specially in Cryptomeria japonica.[253] The scales of the male catkins of conifers likewise occasionally assume the appearance of leaves; this may be seen in monstrous catkins of Araucaria, as also in Podocarpeæ and Cupressineæ (Eichler).

Phyllody of the calyx.—Sepals under ordinary circumstances are so like leaves, that it is not wonderful that they are often replaced by those organs.[254] A singular instance of this has been mentioned as occurring in Cakile maritima, wherein the sepals were found by M. Fournier to be pinnatifid like the ordinary leaves of the plant.[255] The sepals of Ranunculaceæ and Rosaceæ, for example, Rosa, Geum, are particularly liable to this change.

In a species of Geranium recently examined the sepals presented themselves in the form of three-lobed leaflets; so in fuchsias and in Epilobium hirsutum the sepals occasionally are not distinguishable from ordinary leaves (fig. 130). In roses, the change in question is a very frequent accompaniment of prolification (fig. 129). In the peach also this replacement of the sepals is sometimes carried to such an extent, that five perfect, bistipulate leaves occur in the place of the calyx, but when this is the case it usually happens that the pistil is abortive.

De Candolle[256] figures a curious instance wherein the pappus of Podospermum laciniatum was replaced by five linear, foliaceous lobes. A similar change has been noticed in other composites, as in Tragopogon pratense. Engelmann mentions as subject to this hypertrophy of the pappus, as it may be termed, Scorzonera octangularis and Senecio vulgaris. Wigand has observed a similar transformation in a species of Centranthus (Valerianaceæ).

In some cases the phyllody of the sepals has a special interest, as bearing on the question whether what is termed calyx-tube is or is not a portion of the calyx, and whether the sepals are modifications of the blade or of the sheath of the leaf. Thus in the primrose the phyllodic sepals seem to show clearly that the sepals are in that plant of a laminar nature (fig. 131). The so-called calyx-tube of roses is elsewhere alluded to. The leaf-like organs sometimes seen at the apex of a cucumber would seem to support the view that there was really a calyx-tube in Cucurbitaceæ adherent to the carpels. It is also shown in the cut, fig. 132, borrowed from the 'Gardeners' Chronicle,' 1859, p. 654.

Under ordinary circumstances, the sepals may be considered as the representatives of the sheath of the leaf (cataphyllary) or of the blade (euphyllary), the arrangement of the veins being different in the two cases; thus, in the vagina or sheath, there are generally several large veins of about equal size, either convergent towards the apex, or divergent; on the other hand, in the blade, there is usually but one central vein, the midrib, larger than the rest, and the smaller veins come off at a less acute angle, and are more reticulated.[257]

Now, when phyllomorphy occurs in sepals which ordinarily are vaginal, it is obvious that the case is one, not merely of increased relative growth, but also of the appearance or development of an organ habitually suppressed; on the other hand, when phyllomorphy occurs in sepals which usually are laminar in form and nervation, the case is one of unusual growth or hypertrophy, and not of the development of an organ habitually suppressed, so that the amount of change is greater in the former than in the latter instance.

Under normal circumstances it will be found that laminar venation is most common in gamosepalous and vaginal venation in polysepalous calyces. And the same holds good in cases where the calyx is abnormally leafy. The complete leaf development shows itself more frequently among the monosepalous plants than in the polysepalous ones, as shown even in the subjoined list of species. This statement would be more fully verified were it possible to state the frequency with which the condition occurred in individual plants, when it would be found that phyllody of the calyx occurs much more often in individual gamosepalous plants than in polysepalous ones.

Phyllody of the calyx has been most often observed in the following plants:

Phyllody of the corolla.—The petals also are frequently replaced by leaves, though in many of the recorded instances the change has been one of colour only; these latter are strictly cases of virescence. M. Seringe[259] speaks of a flower of Peltaria alliacea in which the calyx was petal-like, while the corolla was leafy as if there had been transposition of the two organs, a very rare, if not unparalleled, instance. In a flower of Campanula Medium, provided, as is often the case, with a double corolla, the outer corolla was slit down on one side, the edges of the cleft being leafy.

The frondescent petals are very often completely disjoined, as in Verbascum nigrum, and Lonicera Periclymenum, in which, moreover, median prolification generally coexists. In the case of Tropæolum majus, the ordinary leaves of which are peltate and orbicular, the petals when frondescent have not the peltate arrangement, but are spathulate, and provided with very long, narrow stalks, so that, in some cases, they are, more properly speaking, enlarged virescent petals than true leaves; in other instances, however, the arrangement of the veins is more like that of the true leaves than that of the petals.

As might be expected, frondescence of the petals is frequently accompanied by other changes of a similar nature in other parts of the flower, and sometimes by the abortion of the sexual organs. Thus, in Actæa spicata, as observed by Fresenius, the petals were replaced by true petiolate, palminerved, lobed leaves, the stamens and pistils being abortive. In Ranunculus the leaves that appear in the place of the petals have no scale at their base, and in Tropæolum the calyx (or receptacle) is free from the usual spur.

The absolute frequency of this occurrence seems to be greatest in those flowers which are normally polypetalous. The petals of these flowers, as a general rule, are more like the leaf-sheaths than the leaf-blades as to their venation, hence it would seem that the phyllomorphic condition in these petals is a manifestation of a greater degree of organizing force than that which occurs in those cases where the petals are normally present in the form of contracted blades or laminæ. (See the remarks in the preceding section.)

Frondescence of the petals has been observed most frequently in the following cases; some, perhaps, were cases merely of virescence, q. v.; see also under Chloranthy, Prolification.