Natural History Review/Series 2/Volume 2/Number 5/On the Morphology of the Female Flower of the Abietineæ

by Robert Caspary

IV.—On the Morphology of the Female Flower of the Abietineæ. (De Abietinearum Carr. floris feminei structura morphologica. 4to. pp. 12. Regiomonti Pr.) By Professor Robert Caspary.^[1]

The structure of the female flower, or inflorescence, of Gymnosperms, has been the cause of more discussion than most questions of morphological botany. In earlier times, the views of botanists on this difficult subject were necessarily uncertain and arbitrary, because no accurate observations had then been made either of the perfect flower or of its evolution. As the study of morphology depends upon the correct knowledge of the taxis of the leaves and branches, the structure of the flowers of Gymnosperms could not possibly be understood at an earlier period. It is not my intention here to enter into details (which may be readily found elsewhere) reading the various explanations which have been proposed; but I think it desirable to state the plain truth as respects the group of Abietineæ, so as to correct previous errors, and to obtain a basis of comparison to which the flowers of other Gymnosperms may be referred.

The true structure of the flowers of Abietineæ was described by A. Braun, as early as 1853, in the following terms, in a note of great importance, though short and modest,^[2] which has been entirely neglected by subsequent writers. "The seed-bearing fruit-scales of the cones of Abietineæ, which lie in the axils of the bract-scales, have all the appearance of one-leaved shoots, but the progressive modifications of form exhibited by these scales in abnormally developed cones of Pinus Larix, prove that each scale consists of two leaves united together." In 1860^[3] he expresses himself in a similar manner, only in more general terms, as to the structure of other Conifers and Cycads. The woody scales of the strobili of Abietineæ consist, according to A. Braun, of two carpels, which originate together, and are the first leaves of an undeveloped bud in the axil of the floral leaf

Before these views of Braun were known to me I was led to the same conclusions, at Bonn, in the autumn of 1858, by the examination of larch cones, which had grown out into leafy branches in the axils of the scale. My observations on these were to the following effect:—

Along the axis of these cones or strobili are inserted linear, elongated bracts, with the woody scales in their axils. The axis does not end with the uppermost scales, but is abnormally prolonged for several inches beyond the apex of the cone. Such strobili have been figured by Richard.^[4] The slender prolongation of the axis differs in no respect from a vigorous annual leaf-branch. It bears linear leaves, of the usual form, each of which has a leaf-bud in its axil. These leaf-buds are subglobose or ovate, and are covered by brown scales (nieder blätter). The two lowest of these scales, which are the most important as respects the true morphological structure of the cone-scale, stand right and left, as in most plants. These are the leaves commonly called the cotyledons of the branches. In these elongated cones there is generally no passage from the woody (seed-bearing) scales of the cone to the leaf-buds. Although I have examined more than 100 such scales, I have met with but few intermediate states explanatory of the true nature of the woody scales. In such intermediate states the cone is not, as usual, shortly ovate, but oblong, and attenuated at the tip, and the woody scales are a little emarginate at the apex. In the scales which appear to pass into the leaf-buds this emargination becomes by degrees more and more deep, till at last, near the summit of the cone, where they are more laxly imbricated, the woody scales are divided, almost to the base, into two obovate or ovate lobes, which are rounded at the apex, or a little mucronate, and are made inæquilateral, by an indentation on the outer side, below the apex. Each of these lobes bears on its inner and upper side, towards the lower margin, the ovate-globose rudiment of an abortive bud. Between the main axis and the bipartite scale I could see no bud. Further up on the axis the intermediate forms are further advanced. The scale is completely bipartite, and the segments are smaller, oblong, subtrapezoidal, obliquely truncate above, with rounded angles, and often wider upwards. As these scales present not even a trace of an ovule, they can no longer, with propriety, be called carpels; but it is most important to observe, that, between their segments and the axis, a leaf-bearing bud, covered with scale-like leaves, is developed. Still higher up on the axis the segments of the woody scale are smaller and more distant from, one another, occupying, by degrees, a more and more lateral position with respect to the leafy bud developed between them and the axis, and approach gradually more and more in size, position, and shape to the two lateral scales of an ordinary leaf-bud, so as at last to pass completely into it.

It is thus clearly proved that the woody scale of the larch cone consists of the first two lateral scales (squamiform leaves) of an undeveloped leaf-bud placed in the axil of the bract which supports the woody scale, these two lateral scales springing in a united state from the outer side of the axis and ascending obliquely. This structure of the woody scale of the larch cone, and consequently of all Abietineæ, is so clearly and irrefutably shown by these monsters, that all other opinions on the morphology of the scales of Conifers are thereby demonstrated to be erroneous.

Among these errors may be mentioned the opinion of M. Baillon,^[5] laid before the French Academy on the 9th July, 1860. Baillon, after examining the evolution of the flower of Taxus baecata L., Phyllocladus rhomhoidalis, Rich., Torreya nucifera, Lieb., Thuja, Pinus resinosa, Salisburia, Gingko, Sm., and Cupressus, arrived at the opinion that the organs which R. Brown regarded as naked ovules are flowers reduced to a pistil, formed of two carpels, and enclosing one orthotropous ovule reduced to a nucleus; and that these very simple flowers are never inserted on a leaf or "bract" (or rather "carpel"), but always on the axis, on which they are sometimes terminal and sometimes lateral; and further, that the cupule or aril of Taxineæ is a dilatation of the axis, "commonly called a disk."

Now what are the reasons which lead M. Baillon to regard the naked ovule of Robert Brown, and almost all recent botanists, as two united carpels? He states that the first developed part of the flower (or what is usually called naked ovule) of all Conifers consists of two small tubercles, opposite to one another, and shaped like a horse-shoe, exactly resembling the carpels of Amarantaceæ, Chenopodiaceæ, &c., in the first stage of evolution. From this resemblance, he regards these tubercles not as the integuments of an ovule, but as carpels, and states that their apices afterwards form two equal or unequal styles. The nucleus of the ovule, according to him, appears after these carpels. This period of evolution is described in detail in Pinus resinosa, and illustrated by figures. As regards that part of M. Baillon's opinion which relates to the more tardy appearance of that which he calls the ovule, his figures do not show it to be the case; but, on the contrary, in t. 1, f. 10, in which the earliest rudiments of the "carpels" are shown, the ovule is also represented, so that M. Baillon's words^[6] are contradicted by that figure. Baillon's statements regarding the evolution of the flower of Conifers are confirmed by M. Payer,^[7] who seems to have examined Pinus and Cupressus chiefly. Payer, however, speaks in such a manner of the time of appearance of the "ovule" and "pistil" that it is doubtful which of the two he considers to appear first; but whatever his opinion may be, he, at all events, does not confirm M. Baillon, for he says "the flower appears in Cypresses and Pines as a little protuberance, on each side of which arises a little ridge resembling exactly . . . . . a very young leaf"

The priority of origin of the outer covering (carpels of Baillon), or the central body (ovule of Baillon), should by no means be neglected, as its determination may assist in fixing the nature of both. For if the central protuberance appear first and the external envelope later, the central protuberance is an ovule, because the nucleus appears before the integument; but, on the other hand, if the exterior envelope appear first and the central swelling later, the body is a pistil, because the carpel always appears earlier than the ovule.

This mode of discovering the nature of the parts fails only in cases where a single ovule appears to be a direct continuation of the axis, as in Rheum,Polygonum, &c. because in these cases it is impossible to decide upon the instant of time at which the apex of the axis becomes changed into the nascent ovule or its nucleus, Baillon,indeed, mentions that the ovules of Conifers arise from the axis; but if the contrary view be established, the test of priority of origin may certainly be applied to the determination of the nature of the different parts of the organ under consideration.

Early in January, 1861, I examined, for the purpose of testing M. Baillon's statement, the female flowers of Thuja orientalis L., Taxus baccata L., Cupressus sempervirens L., Callitris montana, Juniperus communis L., J. sphærica Lindl., J. Sabina L., J. virginiana L., and Pinus Larix L. The climate of Regensburg not being hot enough to enable Gingko biloba, Phyllocladus or Torreya to flower, even in the greenhouse, I regret not to have had it in my power to examine more than a very few of the species on which M. Baillon's observations were made.

With the exception of the Larch, the flowers of all the plants which I examined were almost fully, or at least half, developed; but even in this state of advancement I was led to doubt the accuracy of M. Baillon's statement, that the outer covering (or integument of authors) consists of two carpels. For when two carpels are present, two separate apices (styles, Baillon calls them,) may be expected to be visible; and, in fact, all Baillon's figures of the adult organs in question show two lobes or spices, as in the figure of Pinus resinosa t. i. f. 23, &c., Thuja orientalis, t. ii. f. 17, Cupressus sempervirens,t. ii. f. 20, 21, Phyllocladus rhomhoidalis, t. ii. f. 24, Taxus baccata, t. ii. f. 14, 16. Except, however, in the Yew, in which I found the micropyle to present the appearance of an arched or more rarely straight fissure, the ends of which are opposite to the two highest leaf-scales (bracts), the margin of the organ in question (Baillion's pistillum) was not, in the plants I examined by any means constantly two-lobed, and in the Jumpers I never observed it to be so. The margin of the "pistillum" of Juniperus sphaerica, which appeared fully developed, was invariably entire, and formed by a circle of ten or eleven cells. In the other species of Juniper it was generally obliquely truncate, and in the same species, nay even in the same specimen, it was at one time irregularly sinuate or repand or toothed, at another emarginate on one side or perfectly entire. In Callitris montana the orifice was very wide and surrounded by about twenty cells, and its margin was either irregular or repand, or 3-4-toothed, or quite entire. I never saw it two-lobed. In Thuja orientalis and Cupressus sempervirens, in which Baillon always figures it as two-lobed, I found it occasionally so, but more frequently the orifice was irregularly sinuate or lobed, irregularly crenate, or even quite entire. Richard, too, thus describes the organs of certain Conifers, e.g. Pinus Cedrus,^[8] "margin unequally and irregularly cut into 2-5 segments, which are irregularly erose toothed or repand;" and Pinus balsamea,^[9] "limb longer on one side and slightly divided at the margin into two or three somewhat unequal lobes." These descriptions and figures throw still more doubt on the existence of Baillon's "two carpels." It was, however, in Pinus Larix, in which I fully studied the evolution of the cone-scales, that I acquired a complete conviction of Baillon's error. In this plant, what Baillon calls the ovule, appears first in the form of a convex, almost hemisphærical boss, around which, some weeks later, the integument is produced, not under the form of two distinct horseshoes, but of a complete ring, uniform in height all round. I tried in vain to find any indication of a double origin. It is impossible to consider the floral organ of Pinus Larix as anything else than a nucleus surrounded by an integument, that is, an ovule; and as it is incredible that the integument of Pinus Larix should, from the first, be a regular ring, while that of the other Conifers examined by M. Baillon presents, in its earliest condition, the appearance of two horse-shoes, the observations of MM. Baillon and Payer appear to me more than doubtful.

Were it however the case, that in some Conifers this integument originated as two distinct tubercles, it would by no means necessarily follow that these two tubercles indicate the presence of two organs of distinct origin, not referable to the integument of the gemmule. For:

1. Two-lipped integuments are occasionally met with, which no one regards as two distinct carpels. Thus in Polygala comosa the outer coat of the ovule is produced obliquely upwards and subcucullate, and is divided by a deep fissure into two lateral lobes. Payer makes no mention of these^[10] in Polygala speciosa, though he figures^[11] the ovule-coat of Tremandra verticillata as two-lipped, which is only the case at a late stage of the development of the ovule. The period at which the lips appear seems, however, of little consequence.

2. Other organs certainly exist, which, though single and not composed of two united together, do yet, at their first appearance, show two distinct apices, as, for instance, the stipules of Victoria regia and Euryale ferox, whose evolution I have examined, and the upper palea of grasses which Payer himself describes and figures in Briza media, Panicum aduncum, Triticum monococcum, Ehrkarta panicea and Stipa juncea.^[12]

3. There are certain ovules whose coats sometimes originate equally all round, while at other times, in the same species and even in the same ovary, they are visible on one side earlier than on the other. This I have noticed in Berteroa incana and Tklaspi arvensis, in whose ovules the lower part of the cylindrical nucleus is thicker than the upper, and the two integuments arise seemingly both at once from the lower thickened part, at one time all round and at the same height, at another time unilaterally. As in these cases, true ovule-coats are developed on one side first, and not equally all round, it does not seem unfair to infer the possibility of their appearance in two distinct places or by two gibbi. The alternation of the two lips of the ovule-coat of Taxus, with the two uppermost bracts, may be understood to depend on the existence of more ample room for development on the two sides where there are no bracts, than on the other two, where the bracts come in contact with the ovule. It is well known, that organs increase most in size and vigour in those parts which are free and not interfered with by other organs, while they are weaker and smaller, where they are pressed on by neighbouring organs and deprived of nutriment. In the Abietineæ this may be the cause of the bilobation of the ovule-coat, as the two teeth in Abies excelsa, for instance, are on those sides of the ovule which are not pressed on either by the axis or the scale.

The second part of Baillon's proposition, regarding the flowers of Conifers, is that they always arise from the axis and never from a leaf or bract, or rather carpel. This is shown to be erroneous as to Abietineæ, at least by the monstrous larch described above. It is also excellently refuted by Baillon and Payer's own observations on Pinus resinosa. As described by Baillon, the scales of Pinus resinosa are developed in the following manner. The scale appears first as a small, dorsally compressed, broad boss in the axil of the bract. From the first boss spring three others, one central and two lateral The lateral bosses become broader, assume the form of auricles, cohere externally, and, increasing mainly in width, are gradually converted into an obliquely ascending lamina, the scale itself, which bears a little above the middle in the median line, the subcentral boss, "the organic apex of the axis," which axis produces no more appendages, increases very little in size, and in the adult state, presents the form of a hook bent inwards and downwards. On the upper surface of the lateral wings, towards their lower margin, which is turned towards the primary axis, the ovules are, according to Baillon, produced at a later period.

From this description it is evident that three distinct organs, all differing in period oi origin, can be distinguished, each of which is developed from that immediately preceding it.

1. The axis which originates in the axil of the bract.

2. The ear-shaped organs, which spring laterally from this axis, and are called by Baillon the two lateral lobes. These ascend obliquely and form the greater part of the scale, but are so situated with respect to the minute axis, as manifestly to exhibit the character of appendicular organs. They form two nearly right angles with the ascending axis, and spread out laterally and almost horizontally, so that no one who has learned even the elements of morphological botany, can help recognising them as leaves, and as the primary and only leaves produced on the evanescent axis.

3. From the two lateral organs spring those third in order, namely, the ovules.

Now it is certainly wonderful, but it is not the less true, that Baillon and Payer, failing to distinguish the second organs (the lateral leaves) from the first, though Baillon's description is sufficiently accurate, have confounded both together, and considered them to be a single organ, called by Payer a flattened form of the peduncle; thus rashly following Schleiden, (who, more than twenty years before fell into the same mistake, of describing the axis and its primordial leaves as a simple axis), and Mirbel,^[13] who 46 years before confounded these three very distinct kinds of organs under the common name of peduncle.

Payer further says,^[14] that "this flattened form of peduncle does not surprise those who are aware of its existence in the branches of several plants, such as Ruscus, Xylohylla, Phyllocladus, &c." No one, however, but a tyro in morphology, would confound the scale of Pinus resinosa, on whose upper surface, almost in its middle, the growing point rises as the hooked apex of an evanescent axis, utterly distinct both in position and direction, from the morphological apex of the lamina of the proper scale, with the flattened branches of Ruscus, &c., whose withered growing point occupies the very apex of the lamina, and in which no trace oi appendicular organs is found below the growing point.

Baillon, in a somewhat impressive manner observes, after stating some opinions of others on the structure of the flowers of Conifers, that "the new modes of observation afforded by the study of organogeny, may with propriety be applied to the verification of these opinions." M. Ballion may learn, from the mistakes into which he has been led by the employment of a method which he and Payer alone imagine to be new, that the different grades of evolution of an organ, cannot be understood without an accurate knowledge of the nature of the axis and its appendages, and of the relations which exist between them. M. Baillon, however, hardly knows the elements of morphology. How, for instance, does it happen, that, at the present day, he uses the term alternate,^[15] which was thus applied a century ago, to describe the arrangement of the bracts of the female flowers of Conifers?

Dr. Lindley,^[16] who considers the scales of pine cones to be carpels, (that is, leaves), refers to a cone-like gall of Pinus abies, figured by Richard,^[17] which he mistakes for a cone, and in which he regards the scales as being changed into the form of the acicular leaves of Pinus abies. Baillon has been led by Lindley into the same mistake, of regarding this gall as a cone, and only differs from Lindley, so far, that he thinks it is not the scales but the bracts which are changed into leavee.^[18] Had Baillon read the passage in Richard, to which he refers, he would have seen that Richard correctly regarded the gall as a leafy branch, changed by the attacks of some insect into a false cone. Degeer^[19] describes the insect by which these galls are made, (Chermes abietis, linn.), and figures it and its gall.^[20] He says, "those who have no accurate botanical knowledge, may readily mistake the galls for fir-cones and fruit." Kaltenbach^[21] says, in like manner, "that these galls closely resemble fir-cones, and may readily be confounded with them by ignorant people."^[22]

From the observations given above, it is certain that the flowers of Abietineæ, consist of naked ovules rising from a carpel, and not of pistils springing from an axis. It has been almost universally acknowledged by authors, from the time of Richard down to that of Baillon,^[23] that the flowers of Conifers and Cycads, are almost uniform in structure, following the same laws, with very trifling differences. It appears, therefore, probable that the ovules of all Conifers, Taxus included, are borne on carpels and not on the axis, though at first sight this appears incredible. I shall return to this subject elsewhere.

 

---

 

1. Communicated by Dr. T. Thomson, F.R.S. An abstract of Dr. Baillon's views, referred to in this paper, is given in the Nat. Hist. Review, Vol. I. Bibliography, p.92.
2. Individuum der Pflanze, p. 65.
3. Ueber polyembryonie und Keimung von Cœlebogyne, p. 243.
4. Mémoires sur les Coniferes et les Cycadées, 1826, t 13, f. 9, fig. repetita in Seringe. Eléments de botanique, 1841, t. xiii. f. 12; et Decandolle, Organographie végétale, 1828, tab. 36, f. 3.
5. Recueil d'observations botaniques, t. i. Paris, 1860.
6. l.c. p. 7. "Ce qu'on voit apparaitre d'abord de la fleur femelle c'est une paire de petites feuilles carpellaires en forme de fer à cheval."
7. In Baillon's paper, l.c. p. 17, et seq.
8. Richard Mém., p. 63, t. xvii. no. 1. f. D.
9. l.c.p. p. 76, t xvi f. L.
10. Organogenie, t xxxi. f. 39.
11. l.c.t. xxix. f. 31, 37.
12. Organogenie, p. 701, et seq.
13. Elémens de physiol. végétale, 1815, i. p. 347.
14. In Baillon's paper, p. 20.
15. l.c. p. 6.
16. Veg. Kingd. p. 227.
17. Mém. t. xii.
18. I. c. p. 11.
19. Geschichte von Insekten, deutsch von Götze, iii. p. 66, et seq.
20. T. viii. f. l—29.
21. Monographie der Familie der Pflanzenläuse, p. 202.
22. I may further refer, for information about these galls and the insect which produces them, to Burmeister, Handbuch der Entomologie, ii. 1. abtheil, p. 90, and Koch, die Pflanzenläuse (aphider), p. 317, where the insect is well figured at f. 387 and 388.
23. l. c. p. 11.