[1] Guettard and GÉrard de Villars. Bernard de Jussieu also, who demonstrated that starfish can regenerate.

[2] An annelid of unknown species.

[3] This statement of Spallanzani’s I interpreted incorrectly (’99), thinking that he obtained a two-tailed form as had Bonnet.

[4] There is some doubt in regard to this statement of Spallanzani’s. In a letter to Bonnet he denies that this takes place in the earthworm.

[5] Spallanzani refers to the work of Ginnani, Vandelli, Vallisneri.

[6] He found that the legs of the tadpole of the frog, and of two species of toads, also have the power of regeneration.

[7] These experiments on the earthworm are in the main taken from my own results (’95) (’97) (’99).

[8] Lunularia vulgaris.

[9] Gesammelte Abhandlungen, No. 27, p. 836.

[10] The fact that the piece does, or does not, take in food has no bearing on the question, since many animals that do not feed while the regeneration is going on produce new cells to form the new part.

[11] These two kinds of regeneration are post-generation and regeneration proper. The distinction that Roux attempts to make between these two processes is to a certain extent artificial and rests at present on a very unsafe basis, at least in so far as the post-generation of the frog’s embryo is taken as a representative case of this process. Roux states that in the process of regeneration the injured tissues produce each their like in the new part, while in the process of post-generation of the frog’s egg the new cell-material arises in part from the nuclei and yolk-material of the injured half and in part through the accidental position of the nuclear material of the uninjured half. In order more fully to understand this distinction the original description of the process of post-generation given by Roux in his account of the development of half embryos of the frog’s egg must be referred to. In later papers Roux pointed out that the missing half of the frog embryo, as well as of other forms, may be post-generated without any new material appearing at the open side of the embryo. It is unfortunate, I think, that the original term should have been extended to include these other processes that do not partake of the nature of post-generation as at first defined, but are more like the true process of regeneration as described by Roux.

[12] Ergebnisse der Anatomie und Entwickelungsgeschichte. 1891-1900.

[13] As used in connection with other terms, see his Ges. Abhandl., Vol. II, page 41.

[14] Die Zelle und die Gewebe.

[15] Hertwig’s description of the method by which a piece of hydra makes a new one shows that he did not understand the kind of change that takes place in this animal.

[16] Organographic der Pflanzen, ’98.

[17] This term is used by Driesch to his Analytische Theorie.

[18] Delage, Y. La Structure du Protoplasma, etc., ’95.

[19] The dark red glass was fairly monochromatic; the dark blue let a trace of red light through.

[20] The same difference was found in this form in regard to heliotropism.

[21] PalÆmon and Sicyonia.

[22] The regeneration of the lens of triton may also be affected by gravity.

[23] Driesch does not give in his paper (’99) the position of the hydroids, or the method of the experiment, but I can supply the details given above from a personal communication from Driesch.

[24] Jacobson has shown that the layer of water just above the sedimentary layer at the bottom is poor in oxygen.

[25] “There is thus manifested in the formative force of the tubularia-stem a well-marked polarity, which is rendered very apparent if a segment be cut out from the centre of the stem.” Allman (’64).

[26] The same holds good for the basal hydranth if it arises near an oblique end.

[27] Although it is by no means certain that the results may not be due in part, at least, to injuries to the nervous system.

[28] In normal animals some have the right claw the larger and some the left.

[29] In other plants, fumaria, for example, non-nucleated pieces do not seem to be able to make new starch after using up that which they contain at first.

[30] I have found that the closing in takes place equally well when one per cent of KCl is added to the sea water. This salt has, as Loeb has shown, an inhibiting effect on muscular contractility,—not, however, on amoeboid movements.

[31] Knight obtained similar results in 1809.

[32] VÖchting points out that a parallel case is found in certain conifers. In these there arise from a vertical many-sided main stem whorls of side branches that are symmetrical in one plane. These lateral branches, if cut off and planted, produce new roots and new branches, but the latter are always side-branches, like the parts from which they arise. They never produce a normal main axis. Nevertheless, although these branches cannot themselves produce a main shoot, a callus may be formed at the base of the piece, and from this a new main stem may arise.

[33] A piece suspended in ordinary air dries up without producing any new structures.

[34] Goebel, ’98, page 37.

[35] Examples of this are found in Lilium candidum, Lachenalia luteola.

[36] Delage and Giard give Lessona (’69) the credit for first stating that the phenomenon of regeneration is an adaptation to liability to injury; but RÉaumur first suggested this idea in 1742, and Bonnet in 1745. Delage’s interpretation, viz. that Lessona ascribed this to a prÉvoyance de la nature, has been denied by Lessona’s biographer, Camerano (La Vita di M. Lessona, Acad. R. d. Torino, 2, XLV, 1896), and by Giard (Sur L’autotomie Parasitaire, etc., Compt. Rendus de SÉances de la SociÉtÉ de Biologie, May, 1897).

[37] Whether, having once failed in this way to obtain the snail, the bird or lizard would not learn to make a frontal attack is not stated. Or shall we assume that the tail is all that is wanted?

[38] The Germ Plasm. Translation by W. Newton-Parker, 1893, page 116.

[39] There are no facts that show that this statement is not entirely imaginary. T. H. M.

[40] The italics are, of course, my own. T. H. M.

[41] Fundulus heteroclitus, Stenopus chrysops, Decapterus macrella, Menticirrhus macrella, Carassius auratus, Phoxinus funduloides, Noturus sp., and a few others.

[42] See Newport and Scudder.

[43] Brindley, ’97.

[44] Lepelletur, Nouveau Bulletin de la SociÉtÉ philomatique, 1813, Tome III, page 254; Heineken, Zool. Journal, 1828, Vol. IV, page 284 (also for insects, ibid., page 294); MÜller, Manual de Physiol., Tome I, page 30; Wagner, W., Bull. Soc. Imp. Natural., Moscow, ’87.

[45] The Sarasins have described several cases in Linckia multiformis in which an old arm has one or more new arms arising from it. In one case (copied in our Fig. 38, G), four rays arise from the end of one arm, producing the appearance of a new starfish. In fact the Sarasins interpret the result in this way, although they state that there is no madreporite on the upper surface, and they did not determine whether a mouth is formed at the convergence of the rays, because they did not wish to destroy so unique a specimen—even to find out the meaning of it. There seems to me little probability that the new structure is a starfish, but the old arm has been so injured that it has produced a number of new arms.

[46] For a review of the literature see Brindley, ’98.

[47] I do not know whether this animal was kept long enough to make it certain that the legs do not regenerate.

[48] A statement to the contrary quoted in Darwin’s Animals and Plants under Domestication is doubted by Darwin himself.

[49] The stork and the fighting cocks.

[50] See Darwin, loc. cit.

[51] The more generally accepted results are given in Virchow’s Cellular Pathology and in Ziegler’s Pathological Anatomy. An excellent review of the subject down to 1895 is given in a summary by Ludwig Aschoff in the Ergebnisse d. allgem. patholog. Morphol. und Physiologie, 1895, “Regeneration und Hypertrophie,” in which there are two hundred and eighteen references to the literature.

[52] Nothnagel gives a review of the subject down to 1886 in an article entitled “Über Anpassung und Ausgleichung bei pathologischen ZustÄnden. Zeitsch. f. klinische Medicin.” 1886. Vols. X and XI.

[53] Not, however, from the same litter.

[54] Internat. BeitrÄge zu wissensch. Medicin. Festschrift fÜr R. Virchow, Vol. II, 1891.

[55] Vorlesungen Über allgemeine, Pathologie, Vol. I, 1882.

[56] Handbuch.

[57] Handbuch d. allgem. Pathologie, 1879.

[58] Allgemeine Pathologie, Vol. II, 1889.

[59] Über Endothelwucherungen in Arterien. Beitr. z. pathol. Anat., Vol. VIII, 1890.

[60] Haeckel (1870) first showed, in another medusa, that pieces produce new medusÆ.

[61] In rodents, however, the incisors continue to grow throughout the life of the animal.

[62] If the young worm is fed the new part becomes almost as broad as the old piece, but if the worm is not fed the old part decreases in breadth and the new part does not grow as broad as in the former case.

[63] See Fraisse for literature.

[64] In the figure one double or forked toe is present.

[65] A parallel case is found when a piece partially split in two at the anterior end (Fig. 24) produces one or two heads on each half, according to the extent of fusion of the new material that goes to form the new head or heads.

[66] See Lang (’88).

[67] See Zacharias (’86).

[68] See Hescheler (’97).

[69] The proglottids of the cestodes seem to be an exception, but they are little more than sacs filled with embryos at the time of their separation. How far regeneration may take place in the scolex, or young proglottids, is not known, but it is not improbable that some of the abnormal forms that have been described may be due to regeneration.

[70] Except for the protozoa.

[71] The Fisheries and Fishing Industries of the United States, Washington, 1887.

[72] “The American Lobster,” Bull. U. S. Fish Comm., 1895.

[73] RÉaumur in 1742 records the first observations. Spallanzani also described the process, and many later writers have examined it.

[74] The phenomenon has been observed by Dalyell, Semper, Minchin, and others.

[75] MÜller, Elements of Physiology, 1837.

[76] By Wagner (’87).

[77] For references to the literature on grafting in plants see VÖchting (’84).

[78] In one case they separated only after three months.

[79] This and other experiments were carried out by pushing the pieces on a bristle.

[80] Rand found that when a posterior piece was grafted by its cut, oral end to the side of another hydra that it was absorbed into the stock. In one case it moved down the whole length of the body of the stock and finally disappeared by absorption into the foot of the stock.

[81] Pieces from male and female colonies of the same species also unite.

[82] See Joest’s Fig. 14.

[83] It is not certain whether this is a head or a tail.

[84] Joest states that this new part is a head, as shown by the presence of food matter in the digestive tract of the posterior piece.

[85] The prostomium was misshapen, so that its specific character could not be made out.

[86] It is known that the process of regeneration of the liver takes place especially from the gall ducts.

[87] In one case I observed rhythmic pulsations in a vessel on one side of the neck, in the region above the pharynx.

[88] The figure was drawn fifteen days after union.

[89] Metschnikoff (’86), Herbst (’92).

[90] Eggs without membranes were placed in sea water without calcium, to which a few drops of sodium hydroxide have been added.

[91] The usual interpretation at present is to regard the proctodÆal ingrowth as ectodermal.

[92] In some species the two proliferating regions seem to be in contact above from the beginning (Hepke, in Nais).

[93] This seems to be true for urodeles, but whether it is true for the anurans is rather a question of definition, as I have pointed out in my book on The Development of the Frog’s Egg.

[94] The attachments of the muscles may be the cause of the break in the middle of the vertebrÆ, rather than between two vertebrÆ.

[95] Prodromo, 1768.

[96] Philipeaux, Comptes rendus de l’Acad. des sciences de l’Institut de France, AnnÉe 1866, 1867.

[97] Todd (Quarterly Journal of Science, Literature, and Arts, Vol. XVI), Blumenbach, Treviranus, Von Siebold.

[98] How the tentacles could have gotten into their normal position is not explained.

[99] The foot sometimes pushes out through one of the slits made by the bristle instead of out of the mouth.

[100] I have given elsewhere (The International Monthly, March, 1901) a fuller treatment of the gastrÆa theory from the historical point of view.

[101] It may be pointed out that there may be really several kinds of homology, such as homology due to similar origin of the blastomeres, or to their position, or to their fate, etc. The confusion that has arisen may in part result from the attempt to make homologous parts agree in all points.

[102] That is, one not depending on inheritance through adult forms.

[103] Biologisches Centralblatt, XV, ’95.

[104] A small amount of embryonic mesenchyme may come from some of the ectodermal quartettes of the embryo and produce the branching muscles of the head, but not the characteristic muscles of the trunk.

[105] Cosmos, Vol. VII, p. 388.

[106] King pointed out the fallacy of this argument.

[107] Roux’s earlier experiments in 1885, in which the unsegmented or segmented egg was stuck and a part of its contents removed, the remaining part making a whole embryo, will be considered in another connection.

[108] This had been first discovered by Newport in 1851.

[109] The cross-section C is reversed as compared with the half-embryo B.

[110] This difference is due, I suppose, to the amount of injury that the nucleus of the injured half may have suffered.

[111] The development of isolated blastomeres of the ctenophore egg shows that this need not be the case.

[112] In one case a half-embryo resulted.

[113] The plane of the first cleavage has been shown in two urodeles to correspond, not to the median longitudinal plane, but to a cross-plane of the embryo.

[114] In some cases, especially in sphÆrechinus, even at the eight-celled stage, the blastomeres seem to shift their position, so that a whole sphere of half size is formed.

[115] Hertwig had a year before expressed a similar view in regard to the equivalency of the blastomeres.

[116] A view advanced by PflÜger.

[117] The evidence to show that more than four and certainly more than eight such groups that come from a single egg can produce a pluteus is, I think, insufficient, and the result improbable.

[118] Driesch’s figures seem to show, nevertheless, that the archenterons are proportionately too large.

[119] These may be pieces that were cut obliquely, as Driesch suggests, so that they contain a part of the archenteric region.

[120] Driesch, Hertwig, Roux, Weismann, Barfurth. For review see Driesch (’95).

[121] Bunting (’94) also found that isolated blastomeres of hydractinia make whole embryos.

[122] If the yolk of the dividing egg is partially withdrawn without disturbing the blastomeres, the form of the cleavage may be altered, but a normal whole embryo develops over the smaller yolk sphere.

[123] We offered as a possible explanation in this case that the egg had been cut in two symmetrically with reference to the eccentric nucleus.

[124] These experiments have been quite fully described in my book on The Development of the Frog’s Egg.

[125] Not, however, the supposed action of gravity on the egg.

[126] As stated in my article on “The Problem of Development,” 1900.

[127] According to Roux.

[128] According to E. B. Wilson.

[129] 1897.

[130] Unless it produces a physical change in the structure.

[131] Stevens (’01) has found that this ball of red pigment is ejected from the mouth of the new hydranth.

[132] The importance of this conception is, in my opinion, marred by the fiction of the ferment action of the nucleus; but it should not be overlooked that Driesch avowedly called this a pure fiction.

[133] Not that Driesch supposes this would be the case.