On whorled Phyllotaxis. IV. Early binding whorls

1. A certain number of plant groups with early binding whorls are described. The special results for the particular cases having been dealt with in special paragraphs to which the reader may be referred, the following general results may be mentioned. 2. The early binding whorls in several cases show traces of the original phyllotactical pattern. In the examined cases this pattern almost without exception was a spiral system of the main series, with a divergence between 135° and 144°. As rare exceptions deviating shoots may be found with systems from other phyllotactical series; in an extensive culture of Helianthus annuus these exceptions were collected and investigated. They belonged in majority to the first accessory series, some of them to the first jugate series and a very few to the second accessory series. A real exception to these rules was found in the genera Paronychia and Herniaria, where from the phenomena of the calyx aestivation of the terminal flower it was deduced that the original system in all shoots is a zigzag line, not a spiral; in Rhamnus Frangula the same zigzag is perhaps present. 3. Binding whorls may occur in a radial and in a dorsiventral form. The radial form being primitive, the dorsiventral form is essentially similar but for the supplementary action of dorsiventrality factors, altering the binding whorls to a larger or smaller extent. As these dorsiventrality factors may be lost or may be inactivated, a species with dorsiventral binding whorls may produce a variety with radial binding whorls, and an individual plant with dorsiventral binding whorls may even produce shoots with radial binding whorls. Accordingly the contrast between the two forms has little taxonomic value. 4. For the elaboration of perfect binding whorls the following processes are required: the establishment of the lateral binding into a whorl, the levelling of the whorl members, the equalization of the angles between them, the adaptation of the successive whorls to a certain form of junction, being nearly always alternation, and the elimination of the age differences between the members. These processes may all be executed with a greater or smaller degree of perfection. 5. In consequence of the binding process nearly always whorls are formed. In exceptional cases the binding does not close around the apex, so that partial whorls ensue, as illustrated by fig. 30-32. A more important anomaly is the formation of a secondary spiral instead of whorls, when one end of a horizontal leaf row joins up with a leaf situated above the last leaf of the other end; the binding is then continued in coils winding round the apex. The fact that most plants with binding whorls may produce shoots with biastrepsis as rare exceptions is to be explained in this way. When the successive coils of this secondary spiral are adapted to alternation of the constituent leaves, the secondary pattern assumes the form m + (m+1). 6. In shoots with biastrepsis the secondary phyllotactical pattern may be determined by the observation of the decurrent fibres from the leaves in the stem. In species where the normal shoots are decussate this secondary pattern, as far as our record goes, is always 2 -j- 3. This a> a proof that in such shoots the original system has been of the main series and that the binding has occurred along the SW genetic spiral. In species with polymerous whorls the secondary patterns are 3 + 4, 4 + 5, etc; in many cases however the formulae are less distinctly realized, as the alternation of the successive coils is imperfect, just as the alternation in the polymerous whorls of the normal shoots is. 7. In such cases where either the levelling of the whorl members, or the equalization of the angles between them, or the elimination of the age differences, is not executed to the full, traces of the original phyllotactical system remain. These traces must be distributed in a regular way in the successive whorls, a way which only depends upon the whorl numbers and upon the original system. In systems of the main series alternating dimerous whorls must have a distribution of the peculiar features in antidromous ]/+ spirals, alternating trimerous whorls in homodromous 1/6 spirals, alternating tetramerous whorls in antidromous 3/8 spirals, alternating pentamerous whorls in antidromous 1/10 spirals; in the first accessory series alternating trimerous whorls have their peculiar features in antidromous 1/6 spirals, alternating tetramerous whorls in homodromous 1/8 spirals, etc. These deductions have been fully confirmed by the observation; for the anomalous series see especially § 24. 8. A comparison of the degree of perfection of the adult whorls and that of homologous whorls in the developmental stages shows that the whorl formation processes may have already been carried to completion at the moment in which the primordia begin to bulge out. In other cases they are only partially completed at that moment, so that the remainder has to take place during the visible developmental stages. In a few cases where the transverse metatopies are completed at the moment of bulging out, it is possible to show that the required transverse shifts really have taken place during the invisible stages, by the fact that the median leaf traces only correspond to the shifted insertion place in the upper half of their downward course in the stem, the lower half corresponding to the original insertion places. This is due to the circumstance that the induction of the leaf trace course having taken place during the invisible stages, only the upper half could be shifted with the prospective primordium, whereas the lower half which runs in the domain of other leaf insertions, had to remain in its place. 9. The differences in age of the whorls members, whether eliminated very early or not, of course leave no traces in the adult condition. In a number of species however the axillary buds of different developmental stages, being exposed simultaneously to the morphogenetic forces in the whorl, react differently, so that their development may follow different paths. The difference may only pertain to their vigour, in which case in some plant groups the older, in others the younger buds are furthered, or it may entail a wholly different kind of plastics. In any case the buds of the same kind are distributed in the successive whorls according to the rules indicated sub 7. 10. The formation of binding whorls often forming part of the plastics of only a certain leaf category, other regions of the same shoot may be free from binding. In such cases the original pattern may be traced from the not-whorled into the whorled part and the original order of the whorled leaves may be determined. Shoots, with binding whorls in their lower region only, may have a sudden transition to the not-whorled region (discontinued whorls) or a gradual transition (dissolving whorls). In the former case odd leaves may be left between the whorls and the higher region, in such places as are required by the phyllotaxis of that higher region; in the latter case the transitional region may be particularly instructive through its numerous insufficiences in the whorl formation. 11. In the numerous cases in which the original phyllotaxis of a shoot with binding whorls could be determined in two or more ways, the outcome of these different methods was identical, without any exception. 12. Lateral connation of phyllomcs, being only freely possible in false whorls, occurs in some plant groups with binding whorls, as in Equisetum. If biastrepsis occurs in shoots of such plants, the connation is an additional difficulty for the elongation of the shoot. 13. The explanations, previously given in literature for the origin and the peculiar features of binding whorls and for the existence of bud spirals in the same shoots, based on the principle of rhytm, or on junction phenomena in true whorls, are reviewed and discussed; they are all found to be contra- ■dictory to the facts, with the only exception of the explanation given by Raciborski for the bud distribution in Cupressus. A consideration of the facts to be observed in Cupressus shows that really the case is different from that in Dicotyledons in so far as a sterile whorl between two bud-bearing whorls does not count in Cupressus. Raciborski’s suggestion that the bud distribution in Cupressus is a true cladotaxis, is therefore to be accepted, whereas in the Dicotyledons it is wholly based on phyllotactical processes. 14. In dorsiventral binding whorls the spatial relations between the phyllotaxis and the dorsiventrality factors are usually fixed, at least when the dorsiventrality is clearly pronounced. This relation in dorsiventral decussation amounts nearly always to a diagonal position of the four orthostichies with respect to the ventral and the dorsal sides; in a few plant groups the ventral and the dorsal sides coincide with two of the orthostichies, the other two being lateral. In rare instances the spatial relations may be subject to anomalies or shifts. 15. Under the influence of the dorsiventrality factors a number of special features may arise. The leaves may shift in the longitudinal sense, those at the light side often shifting in the apical direction. The leaves may shift in the transverse sense, often all leaves approaching the dorsal median line. The leaves on the ventral and those on the dorsal side may develop differently (anisophylly), the light side leaves often being smaller, sometimes even rudimentary or aborted. The right and left side of individual leaves may develop differently (inequilaterality). The buds on the dorsal and on the ventral side may get different plastics. In diagonally dorsiventral decussation all similar organs are placed in a zigzag line on the ventral or on the dorsal side. 16. In dorsiventral binding whorls the traces of the original phyllotaxis have usually disappeared. In some species with weakly expressed dorsiventrality such traces may still be present ( Epilobium montanum ) ; in other species the dorsiventral whorls being discontinued, a spiral comes to light (Helianthemum apenninum). 17. The strange fact that the original phyllotaxis of the dorsiventral shoots of Herniaria and Paronychia seems to be a zigzag arrangement, as judged by the calyx aestivation of the terminal flower, a condition which perhaps may also be realized in Rhamnus Frangula, has given rise to the following consideration. If in the shoots ot these genera the dorsiventrality factors are active at an unusually high level in the apex, above the jregion in which the leaf centres are induced, and if by the influence of these factors the dispersion circles at the dorsal side are markedly reduced, the leaf arrangement indeed may become a zigzag, turning at every third leaf.