[13] Since these figures were put on the block, it has occurred to me that the relations would be still clearer, were the primary frond represented as not taking part in these processes of modification, which have been described as giving rise to the erect form; as, indeed, the rooting of its under surface will prevent it from doing in any considerable degree. In such case, each of the Figs. [111 to 117], should have a horizontal rooted frond at its base, homologous with the pro-embryo among Acrogens. This primary frond would then more manifestly stand in the same relation to the rest, as the cotyledon does to the plumule—both by position, and as a supplier of nutriment. Fig. [117a], which I am enabled to add, shows that this would complete the interpretation. Of the dicotyledonous series, it is needful to add no further explanation than that the difference in habit of growth, will permit the second frond to root itself as well as the first; and so to become an additional source of nutriment, similarly circumstanced to the first and equal with it.
[14] How the element of time modifies the result, is shown by the familiar fact that crystals rapidly formed are small, and become relatively large when left to form more slowly. If the quantity of molecules contained in a solution is relatively great, so that the mutual polarities of the molecules crowded together in every place throughout the solution are intense, there arises a crystalline aggregation around local axes; whereas, in proportion as the local action of molecules on one another is rendered less intense by their wider dispersion, they become relatively more subordinate to the forces exerted on them by the larger aggregates of molecules that are at greater distances, and thus are left to arrange themselves round fewer axes into larger crystals.
[15] It is objected that these transformations should be much commoner than they are, were they caused solely by the variations of nutrition described. The reply is that they are comparatively rare in uncultivated plants, where such variations are not frequent. The occurrence of them is chiefly among cultivated plants which, being artificially manured, are specially liable to immense accessions of nutriment, caused now by sudden supplies of fertilizing matters, and now by sudden arrival of the roots at such matters already deposited in the soil. It is to these great changes of nutrition, especially apt to take place in gardens, that these monstrosities are ascribed; and it seems to me that they are as frequent as may be expected.
[16] Since this paragraph was published in 1865, much has been learned concerning cell-structure, as is shown in Chapter VIA of Part I. While some assert that there exist portions of living protoplasm without nuclei, others assert that a nucleus is in every case present, and that where it does not exist in a definite aggregated form it exists in a dispersed form. As remarked in the chapter named, “the evidence is somewhat strained to justify this dogma.” Words are taken in their non-natural senses, if one which connotes an individualized body is applied to the widely-diffused components of such a body; and this perverting of proper meanings leads to obscuration of what may perhaps be an essential truth. As argued in the chapter named ([§§ 74e, 74f]), nuclear matter is, as shown by its chemical character, an extremely unstable substance, the molecular changes of which, perpetually going on, initiate shocks, producing changes all around. In the earlier stages of cell-evolution this unstable substance is dispersed throughout the cytoplasm; whereas in the more advanced stages it is gathered together in one mass. If so, instead of saying there is a dispersed nucleus we should say there are the materials of a nucleus not yet integrated.
[17] This statement seems at variance with the figure; but the figure is very inaccurate. Its inaccuracy curiously illustrates the vitiation of evidence. When I saw the drawing on the block, I pointed out to the draughtsman, that he had made the surrounding curves much more obviously related to the contained bodies, than they were in the original (in Dr. Carpenter’s Foraminifera); and having looked on while he in great measure remedied this defect, thought no further care was needed. Now, however, on seeing the figure in the printer’s proof, I find that the engraver, swayed by the same supposition as the draughtsman that such a relation was meant to be shown, has made his lines represent it still more decidedly than those of the draughtsman before they were corrected. Thus, vague linear representations, like vague verbal ones, are apt to grow more definite when repeated. Hypothesis warps perceptions as it warps thoughts.
[18] Though the subdivision into chambers of the shell does not correspond to the subdivision into cell-units it may still be held that since in the solitary types the subdivision of the nucleus is followed by formation of new individuals which separate, and since in the compound types the subdivision of the nucleus is followed by growth and formation of new chambers, the compound type must be regarded as an aggregate of the second order.
[19] A critic says the question is “what are the forces internal or external which produce union or separation.” A proximate reply is—degree of nutrition. As in a plant new individuals or rudiments of them are cast off where nutrition is failing, so in a compound animal. The connecting part dwindles if it ceases to carry nutriment.
[20] It has been pointed out that I have here understated the evidence of physiological integration. An instance of it among Hydrozoa is shown in Fig. [151], but by a strange oversight I have forgotten to name the various cases furnished by the Siphonophora in which the individual polypes of a compound aggregate are greatly specialized in adaptation to different functions.
[21] Recently Mr. T. H. Morgan has made elaborate experiments which show that Planaria Maculata may be cut into many pieces from various parts and of various shapes—even a slice out of the side—and each, if not too small, will produce a perfect animal.
[22] Since this was written in 1865 there has come to light evidence more completely to the point than any at that time known. In the subdivision of Platyhelminthes known as Turbellaria, there are some, the Microstomida which, by a process of segmentation form “chains of 4, then 8, then 16, and sometimes even 32 individuals.” “Each forms a mouth [lateral] and for some time the chain persists, but the individuals ultimately become sexually matured and then separate.” (Shipley, Zoology of the Invertebrata, p. 92.) Here it should be remarked that the lateral mouths enable the members of a string to feed separately, and that nutrition not being interfered with they doubtless gain some advantage by temporary maintenance of their union—probably in creeping.