1.1. Substance formation and start of the organism

When does a mammalian organism begin? Smith and Brogaard answer: when the process of gastrulation starts, sixteen days after conception. Gastrulation marks a substantial change, according to Smith and Brogaard, because during that process the pre-embryo ‘ceases to be a cluster of homogenous cells and is transformed into a single heterogeneous entity – a whole multicellular individual living being which has a body axis and bilateral.’Footnote ¹⁴ In other words, Smith and Brogaard maintain that human beings do not start earlier than gastrulation, because prior to that process, the cells in the blastocyst are not yet specialised in a way that mark them out as part of a larger contained whole. Thus they do not yet depend on each other in the right way to meet criteria 7–10: forming a unified causal system.Footnote ¹⁵

What is relevant to this paper, however, is Smith and Brogaard's argument that human beings cannot start later than gastrulation. For this they provide two reasons. The first is intrinsic to the foster: any further changes after gastrulation – such as neurulation, the acquisition of organs, hands, feet, etc. – they argue, are not substantial changes that change the foster into a new substance, but changes that the foster undergoes as a unified causal system and human being (criterion 1). The second, which is what concerns us, is extrinsic to the foster: Smith and Brogaard consider but emphatically reject the idea that the foster is part of the gravida. Birth, they claim, is not a substantial change, but merely a transition ‘from one environment into another’, like an ‘astronaut leaving her spaceship’.Footnote ¹⁶ What is their argument? And is it convincing?

1.2. The tenant–niche claim

Smith and Brogaard argue against the view that the foster is part of the gravida by invoking the concept of a niche. A niche ‘is a part of reality into which an object fits, and into and out of which the object can move’.Footnote ¹⁷ Smith and Brogaard maintain that substances s can be inside other substances S, without being part of S, iff S contains a niche that contains s as an occupier or tenant. Gravidae, Smith and Brogaard argue, contain a niche of which the foster is a tenant. Therefore – appearances notwithstanding – fosters are not a proper part of the maternal organism, but instead substances in their own right.

Let's evaluate that. Can a foster move in and out of its niche, the way Smith and Brogaard claim a tenant supposedly can? What a preposterous idea. We cannot take ‘the bun out of the oven’, check it, and stick it back in if it is not fully cooked. (If only!) Birth is irreversible: once a baby is out, it does not go back in – ever. In fact, nobody has even successfully transplanted a foster,Footnote ¹⁸ and even if we did gain that technology, that would not change things. Consider hearts and kidneys; although we have the technology to transplant these, that does not stop us from considering them parts rather than tenants. These organ transplants are possible, but they are exceedingly difficult and risky – and that is precisely because they remove and insert parts of organisms which involves severing major connections. The same will apply should we be able to transplant fosters.

It is clear that, on Smith and Brogaard's initial formulation of the tenant–niche relation, fosters and gravidae do not stand in a tenant–niche relationships. But, Smith and Brogaard do not discuss this initial formulation when discussing the foster. Instead they posit three further characteristics for the tenant–niche relationship that they claim do apply to fosters. These are that niches and tenants (1) do not overlap or have parts in common, (2) do not share an external boundary, and (3) must be separated from each other by some liquid or fluid-filled cavity. As an illustrative example, consider a fish inside an aquarium. Although the fish is inside the aquarium it is not part of the aquarium; rather it is a tenant in a niche. Not only because fish can be moved in and out of the aquarium, but also because (1) fish and aquarium do not share parts; (2) fish and aquarium do not share an external boundary; and (3) fish are in a [water-filled] cavity in the aquarium.

Based on these criteria, Smith and Brogaard give two arguments to convince us that fosters are tenants in a niche. First, and corresponding to the second criterion, they assert that a foster has its own, completely connected external boundary, marked by a physical discontinuity between foster and gravida; the foster is at no point topologically connected to the gravida. Second, and corresponding to the third criterion, they focus on the role of the amniotic cavity as ‘surrounding’ the foster.

1.3. Testing the ‘tenant–niche’ claim: boundaries

Let's assess whether, despite their not meeting the initial formulation of the tenant–niche relationship, fosters should be considered a tenant in a niche according to these new arguments and additional criteria. But before this assessment can be executed, we need to have two further bits of information. First, we must know what the boundaries of the foster are that the tenant–niche criteria repeatedly refer to. Second, we must understand the concept of a fiat boundary.

What are the boundaries of the foster? The answer to this question is less than straightforward, and this is not something that Smith and Brogaard are explicit about. Rather than giving a definite answer, I will identify the three most plausible candidates for delineating a foster. Instead of picking one, I will argue for each of them that Smith and Brogaard's arguments and criteria don't apply, and thus that the foster fails to be a tenant in a niche. These three conceptions of the foster are:

1. 1) ‘Future Baby’ (FB), where the foster only comprises the parts that emerge as the future baby: the (future baby's) body, circumscribed by its skin and stopping at the umbilicus or some way along the umbilical cord;Footnote ¹⁹

2. 2) ‘Baby with Placenta’ (BP), where the foster comprises the ‘future baby’ plus the umbilical cord and placenta;

3. 3) and the ‘Chorionic Content’ (CC), where the foster comprises future baby, umbilical cord and placenta, as well as the chorionic and amniotic membranes and all their contents, including e.g. amniotic fluid.

Before I can examine these three conceptions, it is helpful to understand Smith and Brogaard's concept of the ‘fiat boundary’:

As an example, the closed door to my office marks a real or bona fide boundary between my office and the hallway; when I open my door there is merely a fiat boundary between office and hall.

Smith and Brogaard's core arguments in favour of the fosters’ tenant–niche relationship to the gravida rely on the existence of actual, clear and complete external boundaries. As they themselves state: ‘if the foster is connected to the mother – if, in other words, the boundary between the foster and the mother is a matter of fiat and not of bona fide boundaries – then the foster cannot stand to the mother in the niche-tenant relation’.Footnote ²¹ As the most expedient way of repudiating their claim to about fosters being tenants, then, I will demonstrate that for each conception of the foster its supposed boundary very clearly is one that incorporates a section that is merely a fiat boundary.Footnote ²²

First, consider the Future Baby conception, where the foster has a boundary at the umbilicus or a bit further along the umbilical cord. During pregnancy, this clearly marks a fiat boundary; a physical discontinuity will only appear once the umbilical cord is severed after birth, but does not exist during the pregnancy. The umbilical cord is a clear example of topological connection, and moreover one that marks a functionally and metabolically essential connection. On the Future Baby view, the foster does not stand in a tenant–niche relationship to the gravida.

Second, consider the Baby with Placenta conception. Here the foster has a boundary at the maternal side of the placenta: somewhere inside the spongal maternal deciduous tissue – perhaps at the rough site of the placenta's future separation. This, again is a fiat boundary; after birth, the placenta will detach and there will be a physical discontinuity between placenta and the (previously) pregnant organism. But before birth, there is no boundary here; the placenta is not a clearly defined mass with a smooth surface surrounded by a membrane the way that brains, kidneys or lungs are. Instead, the placenta comprises tissues of maternal and fetal origins and grows direct into (or out of) the uterine wall, just as a tail grows out of the cat. The interface between placenta and the womb is best thought of as a zone, and a zone that is not only marked by the intermingling of fetal and maternal tissue, but also traversed by not one but many functionally and metabolically essential arteries and veins over a large surface. In other words, the placenta is the hallmark of topological connection. A further testament to this is that the wound created upon placental detachment, despite its being a planned and functional biological event, is of such severity that even in our modern age of blood transfusions, it is one of the main causes of maternal death. On the Baby with Placenta view, the foster is not a tenant of the gravida.

Finally, on the Chorionic Content conception, the foster still has a fiat boundary on the maternal side of the placenta, just as it does on the baby with placenta view. For the placenta is part of – and in a sense even outside – the chorion. This view, then, inherits the problem of the baby with placenta view: on the Chorionic Content view, the foster is not a tenant of the gravida.

On none of the plausible conceptions of the foster does the foster stand in a tenant–niche relation to the gravida. On each of these views, the boundary of the foster involves a ‘fiat boundary’ – a boundary that is not marked by a physical discontinuity, but recognized by us for other reasons. The existence of fiat-boundaries mean that the three additional criteria of standing in a tenant–niche relation are not met: foster and gravida do share overlapping parts (either at the level of umbilical cord, or at the level of the placenta/uterine wall); they share an external boundary at these locations; and they are not fully (but only partially) enveloped by a fluid-filled cavity. Nor do Smith and Brogaard's two arguments, which rely on a complete external boundary, marked by a physical discontinuity, and on the ‘surrounding’ of the foster by a fluid-filled cavity, apply. The foster does not have its own, completely connected external boundary, that is marked by a physical discontinuity between foster and gravida; instead the foster is very clearly topologically connected to the gravida at the locus of umbilical cord or placenta, just like a tail is to a cat, your testes to your body, or your kidney to the rest of the organism. Second, the amniotic cavity only partially ‘surrounds’ the foster, which remains topologically connected via a ‘stalk’. Partial surrounding with stalk-connection can not be the sign of tenant–niche status; such morphology is entirely common in mammals: hearts, lung, kidney, brain, pancreas, intestines, and so on are all suspended in fluid filled cavities, but connected by a ‘stalk’.

To conclude, either foster and gravida do not stand in a tenant–niche relation – meaning that instead fosters are part of gravidae – or Smith and Brogaard must have had another conception of the foster in mind than the three I just outlined. Such a conception of the foster – that is free of fiat boundaries – would have to delineate the foster in such a way that no boundary is drawn anywhere between umbilicus and uterine wall. I, for the life of me, cannot imagine what such a conception would be – but I am open to suggestions.

1.4. Interim Conclusion: the part–whole claim

I have repudiated Smith and Brogaard's argument against the part–whole claim: if we take an actual look at physiology, fosters, despite their assertions, do not meet any of their own criteria for being a tenant in a niche. On these grounds I conclude that fosters are part of their gestating organisms, until birth.

Where do we go from here? Rather than accept the part–whole claim, one could reject Smith and Brogaard's specific account of the organism, and look for alternatives. That is beyond the scope of this paper. Even so, we should not assume that other accounts of the organism will easily avoid the present conclusion. The kind of considerations uncovered in this paper, i.e. the intermingling of fetal and maternal tissue at the placenta, the lack of clear boundaries between foster and the rest of the gravida, the sharing of the placental organ, and the evidence of the foster being firmly knitted into the gravida's physical body and metabolic systems, all give evidence that speak in favour of the part–whole claim directly, and may well do so on a range of plausible considerations of the organism. Indeed, I argue in ‘Were You a Part of Your Mother?’,Footnote ²³ at least a first gloss of existing accounts of the organism in the philosophy of biology points firmly towards the claim that the foster is part of the gravida.

If we accept the part–whole claim, what implications does that have for the question of our coming into existence, on an account such as Smith and Brogaard's that connects these questions?

2.1. Attractions

The view that human and other placental organisms start at birth is not, I suspect, immediately appealing. But it does have some attractions. First, it is numerically neat; it is clear how we count organisms. Second, it ties coming into existence to a clear and specific event – birth. This saves us the notoriously difficult and divisive problem of trying to pinpoint an occasion in or impose a dichotomous classification on what is in fact a very slow and gradual but – overall – massively transformative process: conception to embryogenesis to gestation to birth. Third, it is consistent with our intuitive, non-messy view of mammalian organisms in which they are much like we thought they were: clearly demarcated, physically separate individuals from start to finish.Footnote ²⁶ Fourth, it marks out birth as a substantial change. This is attractive because it emphasizes and preserves something that, I submit, too many other views, including popular representations and ordinary language – which often refers to both fosters and babies as ‘babies’ – too easily gloss over (a mistake also made by Smith and Brogaard): birth is a much more substantial event than a mere change of environment, and fosters are not simply ‘babies in tummy's’; there are many differences (internal, structural, functional, relational and topological) between fosters and babies.Footnote ²⁷

2.2. Drawbacks

Even so, the view that organisms start at birth appears to have clear drawbacks. First, it denies that (human) fosters are (human) organisms. Some may be tempted to draw quick moral conclusions from this finding. But I repeat my earlier caution about moral inferences from metaphysical premises. This only seems a drawback if, for example, one believes all of our moral attitudes towards fosters must depend on their being organisms, or if one believes that all parts of our body must be morally similar. Both beliefs need defence. We have no reason at this stage to think that all relevant moral attitudes must track only organisms rather than organisms and fosters. Nor do we have reason to think that all parts of organisms demand the same moral attitude: kidneys and hairs are both parts of organisms, but very clearly deserve different moral and practical attitudes; there is no reason to think fosters wouldn't demand different moral attitudes again. In fact – clearly they do.

Second, a drawback of the view is that it seems so incredibly counterintuitive. It states that the foster does not survive birth, and baby is not the same as the foster that existed seconds before. Because substantial changes block numerical identity on Smith and Brogaard's metaphysics,Footnote ²⁸ and birth is a substantial change, at birth the foster must cease to exist. But that seems unacceptable. Surely a baby does not come into existence out of thin air? Surely the cut of the umbilical cord did not create a human being where there was none before? And on the flipside, what happened to the foster? Did it vanish? Or, if not, where is the corpse?

Such a response, however, misunderstands Smith and Brogaard's ontological view. Substances don't vanish into thin air or defy laws of nature by being conjured up out of nothing. Rather, matter persists to become a different substance; a different thing, when it undergoes substantial change. The house (a substance) ceases to exist when it falls apart into a collection of new substances: a pile of rubble. Thus the house becomes a pile of rubble, but is not numerically identical to a pile of rubble. Similarly, one amoeba ceases to exist when it splits into two new substances: two (new) amoebae. But it is not numerically identical with either or both of them, nor does it leave a corpse. And – to use Smith and Brogaard's earlier claim – a collection of cells, which presumably is a collection of substances, coheres at gastrulation to form a new substance (or so they claim): an embryo. The embryo, on their view, is not numerically identical with the collection of cells. In all these cases there is no corpse – and nothing was conjured up out of thin air. Matter persists, but it instantiates different substances, and substance(s) were transformed into other substance(s).

Although reading the claim in such a context takes away some of the counter-intuitiveness, I don't think it is quite satisfactory. For the worry is not just that it is strange that fosters vanish and babies appear out of thin air – which is a worry we can now set aside – the worry is also that the baby, intuitively, seems to be numerically identical with the foster. Surely newborn babies were fetuses only seconds before? Surely the new mother is holding in her arms the very thing she was pregnant with? And surely dolphins gestate dolphin-organisms that then come out as dolphin-babies? Indeed many philosophers are explicitly committed to the view that (human) organisms were once fetuses.Footnote ²⁹ The beginning at birth view is committed to denying this.

There are two ways of getting around this worry. One is to claim that this intuition in fact tells us that the human being or person persists, and then to deny that the human being is an organism. For example, one might think that the foster is not numerically identical to the baby, but that the human being that is first, say, constituted by or co-located with the foster, and then constituted by or co-located with the baby, persists throughout. But not only does such a view invite objections to the metaphysics involved,Footnote ³⁰ it is simply not plausible that that is what drives the intuition. Remember that we are talking about all mammals here; surely a mouse baby was a foster just as much as a human baby is.

Second, one might attempt to get around the worry by thinking that ‘organism’ is not a substance-sortal, but a phase-sortal, like ‘puppy’ or ‘adult’. Thus, just as my dog has been the same individual throughout its life, but went from being a puppy to an adult to an elderly dog, it went from being a ‘foster’ to being an ‘organism’ at birth. But that clearly fails to capture the spirit of Smith and Brogaard's view. They identify themselves as substance metaphysicians, giving an account of organisms as substances, and very clearly indicating that they are interested in the coming-into-existence of the thing ‘human organism’.

Even so, one might think that this captures something else: the spirit of what we are after more generally. We do want to track and recognize the thing that does appear to persist during birth, whatever that is: the thing that was a foster before and is a baby now; the thing that people adoringly look at on a prenatal scan and whose picture they nowadays share as their baby's first picture. Might we, then, consider modifications to the account of the organism that Smith and Brogaard offer so that we recognise that the foster and baby are the same thing – the same individual offspring? I think we should.

3.1. Revising Smith and Brogaard's account

Can we revise Smith and Brogaard's account of the organism such that a substance survive the transition from being a part to no longer being a part? Whether substances can or cannot be part of other substances is a live subject of debate – and I cannot cover the relevant general metaphysical arguments here.Footnote ³¹ But at least one reason why this might be explored is the following: although Smith and Brogaard's explicit criteria make it very clear that substances can't be part of other substances, other parts of their article are not consistent with that and suggest that parts of substances can be substances in their own right.Footnote ³² Perhaps, then, the idea that substances can be part of other substances was on their mind all along.

I propose the following revisions to Smith and Brogaard's account of the organism:

1. (2) A substance cannot be a part of another substance/organism

   becomes

2. (2*) A substance can be a part of another substance/organism

In addition, if fosters are to be organisms/substances, the following criteria will have to be dropped:

But further modifications are necessary still. Some criteria are only met by fosters to a rather limited degree; fosters certainly depend on internal homeostasis (8), but they can do relatively little to maintain it: temperature, for example, is regulated by the gravida. Similarly, some membranes in fosters protect it against some causal influences (9) – but so do membranes in our body everywhere. Consider for example the blood-brain barrier, or membranes between organs and peritoneal cavity. The true protection for fosters against the outside world, however, is the maternal skin. Finally (10) mechanisms for self-repair and re-balancing are limited: no foster can repair the amnion, placental rupture tends to be fatal, etc – and throughout it all the foster crucially depends on continued ‘support’ of the gravida. Therefore, these criteria have to be toned down.

These revisions, however, come at a cost. First, we have to significantly revise our pre-theoretical concept of organism – and, with that, some of the virtues that that concept had. Second, it becomes difficult to distinguish organisms and organs, and, third, it is even less clear when organisms start, or what their boundaries are.

3.2. Counting Problems

One advantage of the traditional view of individuals is that they are countable. And the unrevised view of organisms, where organisms start at birth, makes for straightforward counting indeed – as we recognise throughout our social and legal practices. The revised metaphysics, by contrast, may create counting problems. In Sameness and Substance, David Wiggins writes of the pope's crown (which consists of three crowns): ‘there is no universally applicable definite way of counting crowns. […] There is no definite answer, when the Pope is wearing his crown, to the question “how many crowns does he have on his head?”’.Footnote ³³ If Wiggins is right, about crowns, then the same may apply to pregnant organisms: there is no definitive answer to the question (when pointing at a pregnant organism at any point in her pregnancy): ‘how many organisms are there?’ This means that there is no definite answer, at any point in time, to the question: ‘how many human organisms exist right now?’ That seems, at the very least, unattractive.

3.3. Distinguishing organs and fosters

A second problem is that, with restrictions on parthood and complete external boundaries gone, it is not longer just fosters that meet the ‘new’ criteria for being an organism; organs are too. Take, for example, a kidney. A kidneys (1) can undergo change (2) must stay numerically the same or cease to exist (3) has spatial parts, [(4) like the foster doesn't have a complete external boundary], (5) is internally connected, (6) is an independent entity,Footnote ³⁴ (7) like the foster, is mostly (but not completely) covered by an external membrane that (9) serves as a (sort of) barrier to the rest of the organism,Footnote ³⁵ (8) depends to a degree on internal homeostasis that it has some role in maintaining,Footnote ³⁶ but also depends on the rest of the organism for this, and (10) engages in some self-repair.Footnote ³⁷ In fact it is not just organs that are organisms on the revised criteria; all sorts of bits of our body seem to be candidates for meeting these criteria, depending on how strictly we apply them: e.g. muscle fibres, fingers, lymph nodes, tongues, the aorta, liver or pancreas lobes, glomeruli, etc. – and even individual cells.

If the revision of these criteria was motivated by the desire to create a more intuitive concept of organism that includes fosters, then this change is hardly successful; that desire is not adequately met by a concept that includes all sorts of human body-parts as human organisms.

How might the criteria be further revised to distinguish between fosters and organs such that the former but not the latter count as organisms? First, one may attempt to distinguish organs and fosters on the grounds that the latter are not essential to the functioning of the larger organism, whereas the former are. That does not work for two reasons. First, fosters are essential for ensuring the reproductive functioning of the gravida; if we respond by arguing that reproductive functioning is not essential, than not only would fosters be organisms and not organs, but so would testes, the ‘corpora cavernosa’ of the penis (responsible for erections), testicles, ovaries, the uterus – and so on. Second, very many muscle fibres, cells and other body bits are not essential; we have a lot of over capacity in various places, including in our kidneys (one is essential – two are not).

Second, one might attempt to distinguish fosters and organs on the grounds that only fosters but not organs can become substances in their own right. But that fails because kidneys, spermatozoa, milk teeth and hair can all become substances in their own right. One might retort that only fosters, but not organs and other body-parts normally become substances. But although that works for kidneys, it does not work for milk-teeth, spermatozoa, hairs, skin cells, cells lining the uterus, etc. – all of which also normally become substances.

Third, one might attempt a distinction on the grounds that fosters but not organs and other body-parts are disposed to become self-standing organisms – or organisms that are also substances. But that at the very least requires more work. First, it is question-begging: if teeth are organisms, on this view, then they are disposed to become self-standing organisms. Second, what about the fosters that in fact aren't disposed to become adult organisms, for example because they lack the capacity to grow beyond a certain point or will certainly not survive birth? And what about the fosters that would have – or would never have – become substances had it not been for medical technology, and/or the mother's exercise of her free will? Thus at least one reason why this position is not easy to cash out is because of the problems of defining ‘normal’ in biology.Footnote ³⁸

I do not rule out that we could have a set of criteria that could successfully discriminate between organs and organisms, and thus between organs and fosters – and a sophisticated version of the third option seems most promising. But such an account needs developing.

3.4. When do fosters begin?

A third issue that remains to be settled is what the spatial and temporal boundaries of the foster are. On Smith and Brogaard's original statement that was already unclear; I noted at least three competitors in the previous section: e.g. the future baby view, the baby-with placenta view, and the chorionic content view. But the unclarity and the possibilities only increase on the revised criteria. Without purported complete external boundaries to do the delineating work, for example, it is not at all clear where gravidae would end or fosters start; whether gravida and foster would not end up having more overlapping parts with, for example, more elements of the mother's anatomy/physiology seen as also part of the foster: parts that it loses upon its birth.

This also re-opens a question that had supposedly been settled: when do fosters begin? One might think that question has been answered; with constraints on maximality and complete external boundaries gone, surely it must be Smith and Brogaard's original answer: 16 days after conception. But since the criteria for organisms have now changed, we cannot take that answer at face value – and I shall argue it is incorrect.

Why? Consider the reasons Smith and Brogaard identified gastrulation as the earliest start of the organism. This was because – they argued – only at gastrulation did the start of cell differentiation and lack of splitting possibilities mark out the foster as a ‘single heterogeneous entity’Footnote ³⁹ or a unified causal system. But in doing so, Smith and Brogaard simply assumed the ‘Future Baby’ view of the foster. Gastrulation is when the embryonic disk – the bit that will make the future baby's body – starts folding into a three-dimensional entity. Other parts of the zygote-derived entity, such as the trophoblast, starts differentiating much earlier, however; around 4–5 days after conception (i.e. just prior to implantation) the blastocyst has already differentiated the trophoblastic cells that help it implant in the uterine lining, and will later build part of the placenta. Thus, if we assume a different view of the foster, such as the ‘baby with placenta’ or ‘chorionic content’ view, then the blastocyst forms a ‘unified causal entity’ at least as early as 4–5 days after conception.Footnote ⁴⁰

One might retort that the possibility of twinningFootnote ⁴¹ counts against recognizing a persisting entity at this stage. And, indeed, Smith and Brogaard discuss this. But that – too – is no longer obvious. On a view where organisms can be part of other organisms, and where – as birth illustrates – organisms can have and lose parts that are or become organisms in their own right, it is no longer so obvious that the possibility of twinning undermines functional integration or blocks the existence of a human being. Indeed, it is not obvious that the possibility of twinning has any relevance to fetal metaphysics at all; on the revised criteria it is entirely normal for a organism to have a part that will or has the possibility of becoming a separate organism.

This suggests that, if we revise Smith and Brogaard's account so as to allow for the possibility that organisms are part of organisms, their criteria may locate its beginning much earlier than 16 days; as early as implantation or even a few days before – only several days after conception. But in the absence of a clear view on the boundary of the foster, it is difficult to tell for certain.

And the boundary of the foster will not be easy to delineate. Once a view allows that organisms are part of other organisms, things get messy (which is precisely one of the attractions of a view that does not allow for that option). Consider again the Future Baby View and the Chorionic Content view. Which one is the foster? Which one is the future organism? I don't know the answer, but I do know that once organisms can be part of other organisms, one might offer quite a few different views. Here is just that ordinarily we probably would not consider: the ‘Future Baby-organism’ is the organism that starts at 16 days and – hopefully – survives birth to die in ripe old age. The ‘future baby’ is part of a second organism – the ‘chorionic content’ organism – which starts a few days after conception with the differentiation of trophoblasts, and dies at birth. The ‘chorionic content organism’ is part of a third organism – the gravida – which preceded and – hopefully – survives birth. This view might solve one puzzle: mono-chorionic twins. On the view suggested one could argue that there is only ever one ‘chorionic content’ organism, that has two future baby-organisms as its parts. But I neither advocate nor defend this view; it is merely to illustrate the options that open up.