ABSTRACT
It appears certain that the process of cellular differentiation is an outcome of interactions between the cell nucleus and cell cytoplasm. Differentiation of embryonic amphibian ectoderm involves two fairly distinct phases: during the first short period an inductor (or some intrinsic factor, if an inductor is absent) determines the course of future differentiation in a multipotential cell; during the second, longer interval of time presumably a complex sequence of reactions leads to physiological and morphological differentiation. Little is known about the nature of reactions which take place during the first phase when the cells become developmentally determined by the inductor. It appears that the first step in translation of the inductive instruction in the competent cell is accomplished during the first 2 or 3 h following the treatment with the inductor (Ansevin, 1966). The step is not actinomycin-sensitive (Ansevin, 1965), as was shown by cells that completely recovered after actinomycin treatment (in conditions when it was unlikely that they could have failed to take up some inhibitor). This suggests that transcription of RNA does not have to occur during cell determination, but it does not exclude the possibility that induction removes biological repressors from specific genes which thus become available to direct cellular differentiation along a specific line. Barth (1965) has drawn attention to a striking similarity between the power of several inorganic ions to bind to DNA and the capacity of the same ions to act as embryonic inductors. He suggested that ions might compete with histones for active sites on DNA molecules, and by binding on these sites protect the molecules from the inactivating action of histones.
MATERIALS AND METHODS
Ovulation was induced in females of Rana pipiens, R. catesbeiana and R. palustris by pituitary injections. R. sylvatica females had been caught while in amplexus and contained ripe oocytes in their uteri. The eggs were fertilized with sperm of either the same or different species.
Haploid pipiens embryos were obtained by fertilizing pipiens eggs with u.v.-inactivated sperm of R. pipiens, or, in other cases, of R. catesbeiana. For irradiation, sperm suspension was pipetted into Syracuse dishes, using only as many drops per dish as were necessary to cover the bottom. The dishes were placed under a GE-G8T5 germicidal u.v. lamp and irradiated for 5−10 min with a dose of 39·5 erg/mm2/sec. Haploid condition of the embryos from eggs fertilized with treated sperm was assumed if the haploid syndrome (Porter, 1939) later occurred in nearly 100% of the unused embryos from a given batch of eggs.
Table 1 lists the nucleo-cytoplasmic combinations which were experimentally produced in the course of this study and which characterized the ectoderm tested for its ability to respond to induction by lithium chloride at a concentration of 3 mg/ml in the culture medium.
Genetic composition of tested ectoderm and number of cultures that were prepared from tissues of each nucleo-cytoplasmic combination

Altogether eggs from ovulations of 8 R. catesbeiana, 10 R. pipiens, 4 R. palustris and 4 R. sylvatica were used in these experiments. Ectoderm was dissected from stage 11 gastrulae, or, in the case of the hybrids, from an equivalent stage. Handling of the tissue and preparation of the cultures were similar to those employed by the author in previous work (1965, 1966), according to the method outlined by Barth & Barth (1962). On the average, 12 embryos of the same nucleo-cytoplasmic type were used for a single experiment. About ten cell aggregates were included in each culture dish. The cultures were maintained for 2 or 3 weeks without changing the medium and examined under the microscope every second day. At the end of the culture period they were fixed and stained with hemalum and acid fuchsin and mounted in toto.
RESULTS
It is evident from Tables 2 and 3 that different rates of response to the inductor and different patterns of differentiation characterized tissues with different genetic composition. Pipiens and palustris diploid ectoderm showed a higher incidence of neural cells as a result of induction by the lithium ion than catesbeiana or sylvatica ectoderm after similar treatment; however, ectoderm of the latter two species had lower thresholds for induction to the melanophore level. Hybrid pipiens ♀ × sylvatica ♂ or palustris ♀ × sylvatica ♀ ectoderm had a lower threshold for melanophore induction than either parental species involved in a particular hybrid. Whether the same was true for the cells of the reciprocal cross R. sylvatica ♀ × R. pipiens ♂ could not be stated with certainty, because of poor viability of these cultures; however, in the surviving cultures nerve cells were more frequent than might have been expected, if quantitative relations between cell types had been the same as in the two former crosses, or if sylvatica cytoplasm had been responsible for the outcome. Thus, both the nucleus and the cytoplasm, rather than the nucleus alone, seemed to control the course of cell determination. It should be added that the modified patterns of cell determinations did not merely result from the presence of only one set of pipiens chromosomes, since haploid condition alone did not seem to modify the diploid pipiens pattern of response to lithium induction.
Sequential induction at 25-27°C in the ectoderm of three other species of Rana and of their hybrids

Since for the type of culture which was employed there is no satisfactory method for estimating the numbers of cells of different kinds, and the observer has to rely on visual impressions, great caution was exercised in evaluating the differences between patterns of differentiation in the cultures. For this reason the real differences between cultures of cells of different genetic compositions are underestimated here rather than overestimated, and it is possible that existing less conspicuous differences have been altogether excluded. Unfortunately, cells of the hybrid pipiens ♀ × catesbeiana ♂ died too soon for differentiation to take place, and thus to contribute any information.
The presence of a hybrid nucleus in the cell manifested itself in some other ways. Pipiens ♀× sylvatica ♂ and palustris ♀× sylvatica ♂ gastrula cells, when spread on the glass, showed distinctly larger diameters than those of pipiens or palustris diploid cells at that stage; however, differentiated hybrid cells were of the normal size. Cells of these two hybrids also showed a considerably greater migratory activity and, apparently, an increased adhesiveness to the glass substrate. In the author’s experience almost no culture of homospecific ectodermal cells had developed such spectacular extensive sheets of cells spreading out of the expiants as did the hybrid tissue cultures. Although viability of the cells of the sylvatica ♀ × pipiens ♂ was poor, in several cultures some aggregates or single cells survived and differentiated. There was a conspicuous difference in the length of survival and degree of differentiation of these cells depending on the pair of animals used for this type of hybridization: cells of embryos derived from one pair did fairly well; those of embryos of another pair gradually died off so that, actually, only a single nerve cell survived until full differentiation, and all ectoderm that had come from a third pair died very soon after explantation.
Uninduced by lithium, catesbeiana and sylvatica diploid ectoderm behaved differently in culture from ectoderm of the other two species. Only those cells adhered to the glass substrate which later differentiated as melanophores. The majority formed aggregates which never adhered to the glass substrate; however, cells of these aggregates seemed to undergo differentiation since they digested yolk, developed and then lost cilia, and eventually formed some type of epithelial tissue.
DISCUSSION
The purpose of the investigation was to learn whether the cell nucleus participates in cellular reactions associated with determination in ectoderm. This was done by comparing patterns of response to time-graded induction by the lithium ion in ectoderm of several species of Rana and of their hybrids. As the results showed, there were differences in the level of induction after the same treatment between ectoderm of different species; also, hybrid ectoderm appeared to differ with respect to its inducibility from ectoderm of either of its parental species. The differences in the hybrid pipiens ectoderm did not depend on the presence of only one set of pipiens chromosomes, since haploid pipiens cells did not show these differences. Unfortunately, the complementary test with androgenetic hybrid haploid cells did not seem feasible in view of the limited viability of the hybrid (R. pipiens) × R. sylvatica, and, therefore, was not attempted.
These results are considered as evidence for the involvement of the nucleus in cell determination: in ectoderm of hybrid embryos the presence of a paternal genome of a different species was found to modify the response to induction which would have been characteristic for the maternal species. Furthermore, the way in which the response was modified suggested that the nucleus influences the process of determination itself regardless of whether or not it also earlier influences the establishment of competence, or later participates in postinductive reactions of differentiation.
The last statement requires some comment concerning what types of nuclear influence should be considered to demonstrate involvement of the nucleus in a particular stage of differentiation. It is unknown when competence is determined by the genes. If it were determined in oogenesis, hybridization should not affect competence; if it were established after fertilization, and if, in a hybrid, genes were incompatible for this purpose, one might expect that: (1) competence would not develop at all, and thus that the hybrid cells could not be induced; (2) competence would be decreased in which case fewer cells would respond to induction, or induction would take longer in time to produce the same result; or (3) competence would be increased and a larger number of cells would become induced or induction would be faster. If, however, genetic incompatibility between maternal and paternal factors affected some postinductive stage of differentiation, one might expect total or partial arrest of differentiation. None of these predictions corresponded to the actual results. Hybrid cells responded to induction which excluded the first possibility and, if anything, induction occurred faster which speaks against the second. Since even very short treatments with lithium ions produced only very few neural cells in hybrid ectoderm, the third possibility seems unlikely. In viable hybrid cells differentiation was never arrested, either totally or at some morphological stage of the process, and thus the postulate that the observed effect was due to an interference with a postinductive reaction becomes untenable. This leaves the hypothesis that the nucleus influences the process of cell determination per se. However, it still remains to be answered whether this nuclear ‘control’ of cell determination is exerted more or less directly on some specific genes, or whether it affects determination only in a very indirect way. For instance, it is conceivable that the hybrid nucleus might affect the properties of the cell membrane, thus modifying its permeability to ions. The results of some experiments of Barth & Barth (1966) suggest that, indeed, permeability of hybrid cells was not the same as that of the normal cell surface, and, in the present experiments, increased adhesiveness and migratory activity of hybrid cells also pointed to changed properties of their membranes. However, modified permeability could probably account for speeding or slowing induction, but not for the qualitative differences in its pattern that have been actually observed. The qualitative aspect suggests a more direct role of the nucleus in the process of cell determination.
Some differences in results of the present experiments and of earlier investigations might probably be explained by different techniques (Moore, 1947; Moore & Moore, 1953), or by different temperatures at which induction and cultivation of cells was carried on (Barth & Barth, 1966) or in the case of lower viability of R. sylvatica ♀ × R. pipiens ♂ cells in Barth & Barth’s experiments (1968), by genetic variability of different pairs of parent animals.
SUMMARY
It is known that gastrula ectoderm of R. pipiens, when treated in vitro, displays a time-or-concentration dependent, sequentially appearing spectrum of cell types following induction with lithium ion.
The pattern of response to lithium ion induction of haploid pipiens ectoderm does not differ from that of diploid pipiens tissue.
Ectoderm of two other species of the genus, R. sylvatica and R. catesbeiana, appears to have somewhat different patterns of response to induction with lithium chloride, while the pattern of ectodermal response of R. palustris is very similar to that of pipiens.
In hybrid cells of R. pipiens ♀× R. sylvatica ♂, R. palustris ♀× R. sylvatica ♂ and probably also of R. sylvatica ♀× pipiens ♂ the maternal species response to lithium ion inductor is modified by the presence of a foreign set of chromosomes.
These results are discussed as evidence for participation of ectodermal cell nuclei in the process of cell determination.
RÉSUMÉ
L’influence du noyau sur la détermination séquentielle de P ectoderme de grenouille à la suite d’une induction par les ions lithium
On sait que l’ectoderme gastruléen de Rana pipiens, traité in vitro, présente une gamme de types cellulaires apparaissant successivement, et dépendant de la durée du traitement ou de la concentration utilisée, à la suite d’une induction par les ions lithium.
Le type de réaction de l’ectoderme haploïde pipiens, à l’égard de l’induction par le lithium, n’est pas différent de celui d’un tissu diploïde pipiens.
L’ectoderme de deux autres espèces du genre, Rana sylvatica et R. catesbeiana, a réagi de manière quelque peu différente à l’induction par le lithium, tandis que le type de réaction ectodermique de Ranapalustris était très semblable à celui de R. pipiens.
Dans des cellules hybrides de R. pipiens ♀×. sylvatica ♂ R. palustris ♀ × R. sylvatica ♂, et probablement aussi de R. sylvatica ♀ × pipiens ♂, la réaction de l’espèce maternelle à l’induction par le lithium a été modifiée par la présence d’un lot étranger de chromosomes.
On discute ces résultats comme démontrant la participation des noyaux des cellules ectodermiques dans le processus de la détermination cellulaire.
Acknowledgements
The author is grateful to Dr L. G. Barth, Dr L. J. Barth and to Dr S. Subtelny for valuable comments concerning this manuscript. She thanks also Mrs E. M. Lewis for technical assistance.