11. JUEL, H. O., Die Kerntheilung in den Pollenmutterzellen von Hemerocallis fulva und die bei denselben auftretenden Unregelmässigkeiten. Jahrb. Wiss. Bot. 30: 205-226. pls. 6-8. 1897.

12.

KorSchelt, E., Zur Frage nach dem Ursprung der verschiedenen Zellenelemente der Insectenovarien. Zool. Anz. 8:581-586, 599-605. 1885. 13. LAWSON, A. A., Origin of the cones of the multipolar spindle in Gladiolus. BOT. GAZETTE 30:145-153. pl. 12. 1900.

14. MERRIMAN, M. L., Vegetative cell-division in Allium. BOT. GAZETTE 37:178-207. pls. 11-13. 1904.

15. MOORE, A. C., The mitoses in the spore mother cell of Pallavicinia. Bor. GAZETTE 36:384-388. figs. 6. 1903.

16. MOTTIER, D. M., Beiträge zur Kenntniss der Kerntheilung in den Pollenmutterzellen einiger Dikotylen und Monokotylen. Jahrb. Wiss. Bot. 30:15-50. pls. 5-3. 1897.

17. OSTERHOUT, W. J. V., Ueber Entstehung der kariokinetischen Spindel bei Equisetum. Jahrb. Wiss. Bot. 30:5-14. pls. 1-2. 1897.

18. WAGER, H., The nucleolus and nuclear division in the root apex of Phaseolus. Ann. Botany 18:29-55. pl. 5. 1904.

EXPLANATION OF PLATES III AND IV

All figures except fig. 28 were made with a Zeiss 2mm apochromatic objective and a no. 12 ocular. Fig. 28 was made with the same objective, but with no. 8 ocular. A Bausch and Lomb camera lucida was used for all drawings. In fig. 2 all the plastids of the cell are shown; in other cases only those immediately surrounding the nucleus.

FIG. 1. Enlarged nucleus of spore mother cell in early stage of preparation for division; the nucleolus is conspicuous and the appearance and arrangement of the chromatin indicate the condition of synapsis.

FIG. 2. Spirem condition, showing linin thread loosely wound with deeply staining chromatic droplets at intervals.

FIG. 3. Thicker and shorter spirem; chromatic droplets fewer and larger; lobes of nucleus distinctly rounded.

FIG. 4. Spirem further shortened; chromatic droplets crowded together; the thread appears double.

FIG. 5. Ends of spirem thread, showing that it is double.

FIG. 6. Aggregation of chromatic droplets just previous to segmentation of chromosomes; probably time of tetrad formation; the nucleolus seems to be fragmenting.

FIG. 7. Later stage than fig. 6.

FIG. 8. Equatorial plate stage, showing group of eight tetrads.

FIG. 9. Equatorial plate stage, later than fig. 8; tetrads not so clearly defined; nucleolus fragmenting.

FIG. 10. A group of selected tetrads, showing rings, crosses, Ys, and Ts; a, tetrad resolved into its elements.

FIG. II. Neighboring tetrads of an equatorial plate; in one the fourfold character is clear, while in the other it is obscured.

FIG. 12. Prophase of first division; nucleus many-lobed; fibers over the largest lobe and at other places on the surface of the nucleus.

FIG. 13. Spindle organizing for first division; spindle fibers prominent on one end, approaching bipolar condition.

FIG. 14. Bipolar spindle of first division; nuclear membrane resolving into spindle fibers.

FIG. 15. Oblique side view, metaphase of first division; end of spindle pointed.

FIG. 16. Metaphase of first division, showing one very flat and one forked pole. FIG. 17. Dividing chromosomes.

FIG. 18. Anaphase of first division, showing chromosomes scattered.

FIG. 19. Telophase of first division, showing grouping of chromosomes in rings at the poles.

FIG. 20. Beginning of reconstruction of daughter nuclei at completion of first division; the chromosomes do not lose their identity and no nuclear membrane is formed.

FIG. 21. Metaphase of second division, showing side view of one spindle and polar view of the other; in the side view the poles are seen to be pointed and in the polar view eight chromosomes appear.

FIG. 22. Anaphase of second division, showing blunt poles.

FIG. 23. Telophase of second division, showing nine chromosomes in the polar view of one of the spindles.

FIG. 24. Telophase of second division, showing beginning of cell plate in one spindle and transverse section of fibers in the other.

FIG. 25. Formation of cell plate.

FIG. 26. Completed spores with resting nuclei and separating walls.

FIG. 27. A single spore which has increased in size and has attained its thickened and roughened wall.

FIG. 28. Resting nucleus of elater.

FIG. 29. Segmenting spirem of cell from seta of sporophyte, showing nine chromosomes.

FIG. 30. Remainder of the same cell, showing seven additional chromosomes, making sixteen in all; chromosomes differ in shape from those of the dividing spore mother cell.

FIG. 31. Early telophase of cell from seta; only half of the cell is shown; there are eight chromosomes at one end and seven at the other; the neighboring section makes it evident that the total number is sixteen at each end.

REGENERATION IN PLANTS. I.

CONTRIBUTIONS FROM THE HULL BOTANICAL LABORATORY.

LXXVI.

WILLIAM BURNETT MCCALLUM.

(WITH FOURTEEN FIGURES)

INTRODUCTION.

THE term regeneration has come to be used by most botanical writers with a broad and somewhat indefinite application. Its essential feature, however, is the replacement of an organ or structure that has been removed. This is accomplished in a variety of ways. PRANTL (9) first found and later SIMONS (11) determined more accurately that if the tip of a root be cut off not more than 0.75mm from the end there is a complete restoration of the part removed, a new tip forming out of the tissues at the cut surface. GOEBEL (3, p. 503) has shown that if the young apex of the frond of Polypodium be cut in two lengthwise, the remaining embryonic tissue on each piece will completely reform the half that has been removed. The same is true of the growing point of a fern prothallium, although the older parts are not replaced. These phenomena are quite homologous with regeneration as it occurs in animals. If we cut off the root tip somewhat farther back, however, a new tip is not organized at the cut surface, but behind it one or perhaps more new root primordia are organized, and these take the place of the main root. Or if we cut off transversely a portion of the thallus of Marchantia or Lunularia (12), the tissues at the cut surface will not develop, but there will arise from apparently mature and differentiated cells back of the cut new outgrowths of thallus which again will complete the plant.

If the shoot with all the buds be severed from the root of Taraxacum, new shoots will arise lower down from the mature tissues of the cortex. Many fleshy roots have this capacity, and if cut into a number of pieces each will organize new primordia and develop shoots. If the young stem of Convolvulus, Linaria, and other plants (6) be cut off just below the cotyledons, there will arise on the sur1905]

97

face of the hypocotyl outgrowths which develop into new shoots. These shoots also arise from mature cells which in the normal course of events remain as permanent tissue. Nor is this power of organizing new shoot primordia confined to the stems and roots, but is also possessed by many leaves, as in the well known cases of Begonia,

Bryophyllum, Cardamine pratensis, Tolmiea Menziesii, and many other plants (fig. 1). 1). Many stems, probably the majority, if removed from the root system and kept moist will produce new roots. In a few cases, as in Salix, there may exist on the stem primordia already organized, but in the great majority of stems these are not present. If a portion of the stem of Salix be cut out from the rest and kept moist, there will appear on it both roots and shoots, each arising, however, from buds already laid down. In the axils of the leaves of many annual shoots are very minute bud primordia, which normally do not develop. If the top of the plant be cut off, these at once form new shoots. In our trees and shrubs the buds formed in the leaf axils do not develop until the following year; but if at any time during the spring the tip of the young shoot be removed, a number of these buds, usually those near the top, at once develop shoots.

FIG. I

We have in these cases at least three seemingly diverse phenomena: (1) the part removed is entirely restored by the growth of the cells immediately at the cut surface; (2) there is no growth of embryonic tissue at the wounded surface, but at a greater or less distance from it the organization of entirely new primordia which develop organs that replace those removed; (3) the organ removed, e. g., the shoot, is restored by the development of already existing dormant buds. Between these no hard and fast lines can be drawn, for they all exhibit intergradations, and between the third case-the development of latent buds and normal vegetative growth no sharp separation can

be made, for occasionally in some species, e. g., Salix, the axillary buds on the first year's growth instead of remaining dormant until the following spring will develop at once into shoots.

It will be quite apparent that as regeneration merges so insensibly into ordinary vegetative growth, the necessary limitations as to the use of the term must be entirely artificial. PFEFFER (8) restricts the term to those cases where an organ directly replaces that portion of itself that has been removed; all others he would call mere reproduction. GOEBEL, KLEBS, MORGAN, KÜSTER, and most other writers on the subject, give it a broader meaning, so as to include the replacement of parts or organs, whether by means of entirely new growths, or from the development of latent buds. The advantage in having some general expression to cover all these phenomena, and the fact mentioned by MORGAN, that they all accomplish the same result and are probably due to the same cause, make it a matter of convenience to use the term in its wider application.

A certain amount of confusion has arisen because it has not been kept clear that regeneration is not really different from ordinary vegetative growth. Most plants naturally tend to grow and branch indefinitely, the new members arising usually in definite places, the shoot primordia, for example, in the embryonic parts of the shoot, and the root primordia ordinarily in the younger regions of the root. The fact that this is the general rule has led to an unjustifiably rigid limitation of the origin of new members to specified regions. As a matter of fact, the ability to produce new members is distributed throughout the plant body, and in many even of the higher plants almost any part is able to produce any other vegetative part. Nor is this ability limited to embryonic parts, for in very many plants it is exercised by the older cells, as in the production of shoots on roots of Taraxacum or on leaves of Begonia. That certain conditions are necessary to bring this latent ability into activity does not make it in the least different from ordinary vegetative growth, for the latter also is dependent on definite conditions.

The whole plant body of mosses and liverworts, and many roots, stems, and leaves of the vascular plants have this capacity, and it requires only the proper conditions to become manifest.

In spite of the extensive investigations into this question, ranging