Two difficulties have to be faced by this theory—(1) the origin of the metabolic power of the cells, (2) the reason why the cells arrange themselves so as to form an organism of complex and definite structure. Schwann tries to explain the origin of the "metabolic" action, the analogy of which with the contact-action of colloidal platinum he recognises, by attributing it to the peculiar structural arrangements of molecules. In attempting to account for the harmonious structure of the organism he points to the analogy of ordinary crystals, which often form complex and regular tree-like arrangements; plants in particular resemble these regularly shaped crystal-aggregates.
The whole ingenious theory is offered merely as an hypothesis and a guide to research. It is interesting as one of the most carefully thought-out attempts ever made to give a thorough-going materialistic account of the origin and development of organic form, and it arose directly out of the cell-theory.
Schleiden and Schwann started out from an erroneous theory of the origin and development of cells, which impaired to some extent the value of their results. It was not long, however, before their theory of the origin of cells by "crystallisation" from an intra- or extra-cellular cytoblastem was challenged and overthrown, and the generalisation that cells originate by division from pre-existing cells put in its place.
This was established for plant cells by Meyen, Unger, von Mohl, Naegeli and Hofmeister in or about the forties.[261] Criticism of the Schwann-Schleiden theory from the zoological side was suggested by the study of the segmentation of the ovum—the developmental process in which the multiplication of cells is most easily observed. The segmentation of the ovum was well known to Schwann, for the process had been described in the frog by Prévost and Dumas in 1824,[262] in the frog and newt by Rusconi,[263] and an elaborate study of the process in the frog had been made by von Baer.[264] Schwann indeed suspected that there must be some connection between the segmentation of the ovum and the formation of cells, but he did not realise that the cellular blastoderm of the chick was formed by the division or segmentation of the egg-cell.
Segmentation was soon found to be of widespread occurrence. Von Siebold in 1837 described the process in Entozoa,[265] and in the same year Lovén saw segmentation in Campanularia,[266] and Sars in the starfish and in Nudibranchs.[267]
In 1838 Bischoff[268] observed segmentation in the mammalian ovum, and the whole course of segmentation in the ovum of the rabbit from the 2-celled to the morula stage was carefully described and figured by Barry[269] in 1839. C. Vogt[270] in 1842 described segmentation in Coregonus and Alytes. The discovery of segmentation in the ovum of birds was not made until 1847, by Bergmann,[271] confirmed independently by Coste[272] in 1850. By 1848 segmentation had been noted in Hydra and various hydroids, in acalephs, in starfish, polyzoa, nematodes, rotifers, leeches, oligochætes, polychætes, in most groups of molluscs and arthropods, and in all the vertebrate classes.[273]
The process was at first held to be merely one of yolk-division, or Dotterfurchung, and its details were by most interpreted in the light of the Schleiden-Schwann theory of cell-formation.
The first steps towards a truer conception of the process seem to have been taken by Bergmann, who in 1841[274] called attention to the presence of nuclei in the segmentation-spheres of the frog's egg, and by Bagge in the same year, who observed that division of the nuclei preceded the multiplication of the segmentation spheres.[275] He considered the nuclei to be anucleate cells, and the same view was taken by Kölliker in 1843.[276] Next year, however, in his classical paper on Cephalopod development[277] Kölliker came to the opinion that they were really nuclei. He showed that segmentation was brought about by cell-division, that between "total" and "partial" segmentation there was a difference of degree and not of kind, and that the cells of the body were formed by division of the segmentation spheres. He held, however, that the nuclei multiplied endogenously and not by division. The division of nuclei was observed by Coste in 1846.[278] Leydig in 1848[279] took the necessary step in advance and maintained that the nuclei as well as the cells increased always by division. He was supported by Remak, who in a paper of 1852,[280] and more fully in his monumental Untersuchungen über die Entwickelung der Wirbelthiere (Berlin, 1850-55), proved that in the frog's egg at least segmentation was a simple process of cell-division, initiated always by division of the nucleus.[281]
One point Remak left undecided—the fate of the Keimbläschen or egg-nucleus. It was generally held, even so late as the 'fifties, that the egg-nucleus disappeared just before segmentation began—Bischoff clung to this belief even in 1877.[282] Though Barry had held in 1839 that the egg-nucleus does not disappear in segmentation, J. Müller seems to have been the first actually to prove that it forms by division the nuclei of the first two segmentation spheres. He furnished the demonstration in the egg of Entoconcha mirabilis,[283] and his paper was known to Remak, who could not, however, observe a similar division of the egg-nucleus in the frog. Müller's discovery was confirmed for Oceania armata by Gegenbaur,[284] and for Notommata sieboldii by Leydig.[285]
In 1854 Virchow,[286] previously a supporter of Schwann, crystallised the new views in the famous phrase—Omnis cellula e cellula—and gave wide publicity to them in his classical lectures on Cellular Pathology, delivered in 1858.[287] The new doctrine of cell-formation was also taught by Leydig[288] in his text-book of histology, published in 1857.