The transformation of a zygote into an embryo adheres both to nature and to nurture: not only genetics but also environment determines the outcome. This idea has assumed many forms. A territory in the early sea urchin blastula contains clones of founder cells, and each clone contributes exclusively to one territory (Cameron & Davidson, 1991). Both the lineage of a founder cell, i.e. its nature, and the position of a founder cell, which determines how it is nurtured, contribute to its fate. Fertilisation, the topic of the first Forum, fixes the genes; interblastomere communication, the topic here, regulates gene expression. Blastomeres communicate like any other cell – via ligand-receptor interactions and through gap junctions. Saxe and DeHaan review these mechanisms. The definition of ligands and receptors becomes broadened in this context, and cell adhesions as well as gap junctions enter into the story. In spite of these entanglements, it appears that nature uses the same sorts of mechanisms to get cells to specialise that she uses to keep them talking. Thus, neurons and glial cells signal to one another via glutamate receptors and gap junctions (Nedergaard, 1994). Likewise, we expect neurotransmitters (and neurotransmitter transporters) to help signal differentiation. The biophysicist may ask whether electrical properties also play a role, but that we reserve for another Forum. If gap junctions figure in development as fusion pores that pass small molecules and electrical signals between blastomeres, another parallel suggests itself. Brian Dale asked in the first Forum: How does a spermatozoon activate an oocyte? This question, which concerns gamete communication, has produced two schools of thought and remains controversial (Shilling et al., 1994). Do sperm activate oocytes via contact-mediated mechanisms or through fusionmediated mechanisms? Or do both mechanisms occur, as they appear to in development?

During the past decade growing evidence has suggested that cell-cell communication via gap junctions is crucial for early developmental processes (Warner et al., 1984; Guthrie & Gilula, 1989; Serras et al., 1989). It has been shown that embryos of mice (Kalimi & Lo, 1988), teleosts (Kimmel et al., 1984), insects (Warner & Lawrence, 1982; Ruangvoravat & Lo, 1992) and molluscs (Serras et al., 1989) become regionally organized into restricted domains of junctionally connected cells that share developmental potential. In the mouse gastrula, for example, dye-coupling experiments have demonstrated that cells within a developmental compartment have a high degree of coupling whereas cells across compartmental boundaries have reduced coupling (Kalimi & Lo, 1988). Classic experiments (Townes & Holtfretter, 1955; Steinberg, 1963) demonstrated that as cells begin to differentiate along common pathways, they develop selective adhesion properties and the ability to sort themselves from unlike neighbours (for review see Edelman 1988; Edelman et al., 1990). More recently a multitude of specific cell adhesion molecules (CAMs) have been identified that mediate these processes and trigger the cytoplasmic events that drive further differentiation (Edelman et al., 1990; Albelda, 1991; Geiger & Ayalon, 1992).

It has been shown previously that, in Drosophila oogenesis, potassium ions are important for bioelectric phenomena as well as for other physiological and development processes. In the present study we determined the spatial distribution and activity of the (Na+, K+)-pump and of ouabain-insensitive K+ pumps in plasma membranes of vitellogenic ovarian follicles (stage 10). We used that light micorscopic anthroylouabain method as well as the cytochemical lead and cerium precipitation methods in combination with electron spectroscopic imaging (ESI) and elelctronm energy-loss spectroscopy (EELS). (Na+, K+)-ATPase activity was predominantly observed on the oolemma as well as on the membranes of the columnar follicle cells covering the oocyte, whereas on the membranes of the nurse cells and of the squamous follicle cells covering the nurse cells the activity was vary low. The highset activity of the (Na+ K+)-pump was found at the anterior and posterior ends of the oocute, and this on the oolemma as well as on the membranes of the follicle cells located here. Strong activity of ouabain-insensitive K+-pumps was observed on most of the oolemma (except at the anterior of the oocyte) and on the membranes of some nurse cells located next to the oocyte, whereas less activity was found on the other nurse cell membranes and on the membranes of all follicle cells. The suitability of the differnet methods nurse cell membranes and on the membrances of all follicle cells. The suitability of th different methods used for determining the localisation as well as the activity of K+-pumps is discussed. We further discuss the nature of the ouabain-insensitive K+ pumps and the relevance of the observed distribution of K+-pumps for K+ uptake, extrafollicular ionic current flow intercelluar signalling and other developmental processes in Drosophila oogenesis.

The morphology of the acrosome of spermatozoa an Australian daysyurid marsupial during in vivo fertilisation, as seen with the tranamission electron microscope, is described. In some spermatozoa close to the outer zona surface with largely intact acrosomes, eletron-demse struts between the inner and outer acrosomal membranes could be seen; in addition fusion between the outer acrosomal and overlying plasma membranes was sometimes evident. The zona matrix close to the acrosomal region of spermatozoa with heads party embedded within the zona pellucida was less electron-dense and more filamentous than elsewhere; perhaps indicative of local lytic activety. Nevertheless some of these spermatozoa at least had partly intact acrosomes with local regions of electron-dense bridges between the two acrosomal membranes. Spermatozoa with condensed chomatin were also observed within the egg cytoplsam some of which had membranes, vacuoles or close to the dorsal nuclear surface which could, in part, be acrosomal in orgin. These observations lend some support to the recent view, based on findings from an American species, that acrosomal changes of spermatozoa of some marsupials at the of fertillsation may be somewhat similar to those of eutherian mamals in spite of the considerable differences in gamete structure and design. How generally applicable these findings are to all marsupials remains, however, to be determined.

Dorsal–ventral patterning in the Xenopus egg becomes established midway through the first cell cycle during a 30° rotation of the subcortical yolk mass relative to the egg cortex. This rotation of symmetrisation is microtubule dependent, and its direction is thought to be cued by the usually eccentric sperm centrosome. The fact that parthenogenetically activated eggs also undergo a directed rotation, despite the absence of a sperm centrosome, suggests that an endogenous asymmetry in the unfertilised egg supports the directed polymerisation of microtubules in the vegetal cortex, in the way that an eccentric sperm centrosome would in fertilised eggs. Consistent with this idea, we noticed that the maturation spot is usually located an average of more than 15° from the geometric centre of the pigmented animal hemisphere. In parthenogenetically activated eggs, this eccentric maturation spot can be used to predict the direction of rotation. Although in most fertilised eggs the yolk mass rotates toward the sperm entry point (SEP) meridian, occasionally this relationship is perturbed significantly; in such eggs, the maturation spot is never on the same side of the egg as the SEP. In oocytes tilted 90° from upright during maturation in vitro, the maturation spot developed 15° or more from the centre of the pigmented hemisphere, always displaced towards the point on the equator that was up during maturation. This experimentally demonstrated lability is consistent with an off-axis oocyte orientation during oogenesis determining its eccentric maturation spot position, and, in turn, its endogenous rotational bias.

Production of embryos that are free of tough outer coats facilitates studies that are not possible with embryos surrounded by impenetrable envelopes. This report describes a new procedure for preventing formation of fertilisation membranes in the sea urchin (Lytechinus pictus) model. This procedure involves treating unfertilised eggs with the enzyme alpha-amylase, which cleaves alpha-1,4 glucosidic bonds in the vitelline layer. A major advantage of this method is that it is very well defined and completely controllable with alpha-amylase inhibitor. The results suggest that intact alpha-1,4 glucosidic bonds are essential for vitelline layer integrity required for formation of the fertilisation membrane. Eggs treated with alpha-amylase possessed the same surface lectin receptors as untreated eggs and, as shown by light and transmission electron microscopy, produced healthy, cleaving embryos that were free of fertilisation envelopes.

A study was conducted to evaluate the ability of mammalian oocytes to develop to the pronuclear stage and beyond if injected with freeze-dried sperm nuclei. The rate of development of hamster eggs to the pronuclear stage after microinjection of freeze-dried hamster sperm nuclei was 90%. The pronuclear formation rate of hamster eggs injected with freeze-dried human sperm nuclei was 85%. On the other hand, the rates for eggs injected with untreated sperm nuclei were 84% and 89% respectively. The survival rate of rabbit eggs microinjected with freeze-dried rabbit sperm nuclei was 64%, the fertilisation rate 56% and the cleavage rate 38%. The survival, fertilisation and cleavage rates of eggs injected with non-freeze-dried sperm nuclei were 78%, 45% and 34%, respectively. There was no difference between eggs injected with freeze-dried sperm nuclei and those injected with sperm nuclei that had not been freeze-dried. Of the viable rabbit eggs injected with freeze-dried rabbit sperm nuclei, 6% developed to the 6- to 8-cell stage 62 h after injection.

Isolated seminiferous tubules of rat testis contain considerable urokinase-inhibiting activity. An immunohistological analysis revealed the presence of plasminogen activator inhibitor type 1 (PAI-1) in the basement membrane as well as in the interior of the tubules. Distribution and intensity of the intratubular immunoreactivity depends on the tubules. Distribution and intensity of the intratubular immunoreactivity depends on the stage of the seminiferous cycle. A relatively weak signal is present around elongated nuclei of spermatids at the beginning of chromatin condensation. The signal intensity increases in the course of differentiation until a maximum is reached at stages VII-VIII. In these stages PAI-1 immunoreactivity is localised around the nuclei of the late spermatids as well as along their tails. Spermatozoa in the ductus epididymis also strongly react with the PAI-1-specific antiserum, suggesting that the inhibitor remains associated with the germ cells after spermiation and during maturation in the epididymis. In intact mature spermatozoa isolated from epididymis cauda by ‘swimming-up’ in non-capacitation medium, PAI-1 antigen is localised on the plasma membrane surrounding the head. In addition, in fixed and permeabilised cells the immunoreactivity is detectable in the acrosome and in the tail. Possible functions of PAI-1 in spermatogenesis, sperm motility and sperm-egg interaction are discussed.

After spermatozoa bind to and penetrate the extracellular matrix of the egg, the zona pellucida, they adhere to and fuse with the plasma membrane of the egg. Since sperm–egg fusion may involve membrane glycoproteins and/or carbohydrate binding proteins, we sought to test this hypothesis by challenging sperm–egg fusion in hamster and in mouse with added carbohydrates. In this study, a number of carbohydrate and glycoconjugates were examined for their ability to inhibit sperm–eggfusion. In the hamster, D(+)-glucosamine, D(+)-galactosamine, albumin-bovine-glucosamide and-galactosamide, fucoidan and dextran sulphate inhibited the fusion of spermatozoa with zona-free eggs. The same effects were seen in the mouse, except for the toxic effects of D(+)-galactosamine. These facts suggest a role of carbohydrate binding proteins or glycoproteins in the fertilisation process at the level of binding to and fusing with the oolemma.

As the amphibian oocyte becomes the fertilisation-competent egg an actin-myosin assembles in the cortex which provides for the cortical contraction that accompanies the fertilisation. A number of recent investigations provide data for development of a model detailing the structural changes which should accompany the development of this contractile network as well as the signalling mechanisms which regular assembly and contraction.