DIE SEXUELLE FORTPFLANZUNG DER PENNATEN DIATOMEEN

Publisher Website
Google Scholar
Abstract
SUMMARY: Although the sexual reproduction of pennate diatoms has long been known in outline, it is only recently, with the help of more exact methods, that problems of the mechanism of copulation, of sex differentiation and sex determination, together with various developmental questions, have been dealt with.In the case of allogamy, the pairing of mother cells and the copulation of gametes should be distinguished. Various species form two gametes per mother cell, others only one. The gametes are always morphologically isogamous; when there are size differences, these are not related to sex differentiation, or cannot certainly be shown to be so. When two gametes are formed per mother cell, there is a great diversity in the manner of copulation, due not only to the constitution of the gametes but also to mechanical factors (Section III). These depend on the mode of pairing and on the nature and behaviour of the copulation jelly. In this respect there are certain relationships to taxonomy (Sections VII, VIII).An analysis of the process of copulation makes it appear doubtful whether the different motilities of the gametes are an expression of their different sexual constitution, whether, that is, in the special case of physiological anisogamy, the active gamete is male and the passive one female. The mother cells would then be hermaphrodite and their paired state would merely be the expression of their sexual activity, but not of their difference in sex. The further the analysis is pushed, the greater are the difficulties of interpretation in terms of a general theory of sexuality. This does not refute such a theory, but shows that other criteria should be employed. The old assumption that the two gametes which form one mother cell are of different sexes, but usually do not copulate because of self‐sterility, cannot at once be discarded in spite of its difficulties, since facultative and obligatory automixis occur, in which case the difference in sex of the two gametes from one mother cell is proved (Section V). Nevertheless, there exist yet other difficulties in interpretation, also in part for those species which only form one gamete (Section IX). in that the two valves of one cell are differently constituted and the gamete arises as a result of a ‘differentielle Teilung’ (differential cytokinesis), which, as regards the valve form of the cell, is orientated in a definite way: the gamete always arises on that side of the protoplast which is nearest to the epitheca; the other daughter cell aborts (Section IV). The epitheca must, however, at the same time be the rapheless superior valve. Cells whose superior valve is formed of the epitheca are unable to pair and they remain vegetative. In addition to the various internal and external conditions which determine pairing and copulation (Sections X, XI), there are others, which result in only 50 % of the cells becoming sexual and capable of pairing, so that only 25 % of the theoretically possible types of zygotes can be produced. In Eunotia an analogous differential division takes place during the formation of the single gamete, which, too, is differently oriented in regard to valve form, but this results in no reduction of pairing, because the thecae are built alike. It is the combination of differential division with dorsiventral cell structure which leads in Cocconeis to the result described above.An analysis of the behaviour of the mother cells at pairing leads to certain general rules, which, however, have to be modified in accordance with the organization and mode of life of various species. In Eunotia the physiological pairing reactions can to a certain extent be deduced from the position of the partner (Section X). In other cases, too, it can be shown that behaviour in pairing is determined in a definite manner.In Cocconeis, which forms only one gamete per mother cell, an unusual complication arisesDifferential, unequal divisions are in every case the consequence of a pregamic differentiation of protoplasts in a pervalvar direction, which consists in the development of the one side and inhibition of the other; in this direction there is a polar gradient. Whereas, however, in Cocconeis the favoured side is turned towards the epitheca, in Eunotia it is towards the hypotheca. Indeed, the development is different in different Eunotia species, insofar as it causes growth of the chroma‐tophore lying on the particular side, or in the migration of both chromatophores to that side. It is noteworthy that an unequal differential division also takes place in the formation of inner valves (craticular condition) in Eunotia, which, too, depends on the development of the hypothecal side. Thus, in Eunotia there is a pervalvar polarity, always with the same direction.Not only does the formation of additional valves (inner valves) require a cell division–even if in this case a highly modified one–but also the appearance of the new valves of the primary cells, developed from the auxospores, is generally dependent upon a division. It is true that this division involves a mitosis, which is an unequal division insofar as it only gives rise to a single surviving daughter nucleus, while the other one aborts. Since two valves are always formed, there are always two such defective metagamic mitoses, namely one before the formation of the first valve and the other before the second valve (Section XII). This behaviour is deeply embedded in the organization of diatoms; valve formation always proceeds from a vegetative division, apparently causally connected with it, or from rudimentary division processes derived from it.In another respect it is remarkable that there are species in which all four tetrad nuclei develop further. From the phylogenetic point of view this is a reminiscence of the formation of four gametes. Thus, while Navicula radiosa only produces two gametes, since there is only one cytokinesis, yet one daughter nucleus of each of the second meiotic telophases does not abort in the usual way, but all four daughter nuclei are taken over by the gametes. In this way the zygotes for have four nuclei, and all of the four are physiologically equivalent, i.e. they pair and form two pairs of nuclei: only later does one pair degenerate. N. cryptocephala behaves essentially in the same way. The question remains open as to which nuclear pair degenerates and why. The same question arises concerning the disappearance of one daughter nucleus of the second meiotic telophase, and in the formation of new valves. Here, as always, new facts produce new problems.

This publication has 20 references indexed in Scilit: