CeleryKills

The Illusion of Inheritance

By late March, when thaw sits in the ground longer than it should and alder pollen powders the air, Mark starts looking at heads, not bodies. He is a field biologist by training and habit, which means he collects regularity the way other people collect stories. He notes pedicle diameter, tine angles, the small asymmetries that accumulate until they look intentional. I follow him some weekends and try to keep up with what he takes for granted. What looks like repetition from the trail often feels, under his lens, like a series of choices made by tissue. This man was born for the PNW.

What I did not expect, at least not at first, was that his obsession with antlers would turn into a problem about information. He had measurements, photos, dates, GPS points scribbled in notebooks that always seemed slightly damp. Patterns were there, but they sat in fragments. That is where I entered the picture. He needed someone to move between what he was seeing and something that could hold it without losing the nuance. I had spent years thinking about how information gets structured, what gets recorded and what gets quietly dropped, and I had just enough biological anthropology to understand why his “small asymmetries” mattered. He talked about tines like events. I heard variables. Somewhere in that overlap, we realized we were both tracking the same thing from opposite directions.

It never stayed abstract. We built systems that had to live in the same conditions he worked in, mud, rain, cold hands and bad signal. Laptop first, then something lighter, something that could survive a weekend in a pack. We argued about fields and categories the same way he argued about growth. What counts as a deviation? When does variation become pattern? I pushed him to collect data he thought was peripheral. He pushed back when I tried to over-structure what he insisted was still unfolding. That tension held. And in between database schemas and wet field notes, I got something I had not planned on. Long days watching animals that did not care about our models, a better sense for what persistence actually looks like in the wild, and at least one afternoon where I learned, very quickly, that sitting loosely on a horse is not the same thing as knowing how to ride it.

Growth begins out of pedicles that refuse dormancy. Mark will kneel beside a shed and point to the base as if it were an instrument panel, vascular, responsive, primed. When spring signals arrive, velvet advances and feeds a rush of proliferation underneath. He avoids the language of construction. He calls it patterned emergence, because the tissue does not assemble so much as resolve itself under gradients you cannot see directly but can infer from the way branches declare themselves (Goss, Deer Antlers: Regeneration, 1983; Kierdorf and Kierdorf, Antler Regeneration, 2011). I hear him say this and then watch a stag in a wet clearing, antlers soft and warm, and I do not think scaffold. I think ongoing argument.

Mark’s work becomes most legible at the moment of interruption. He has records of velvet injuries from fences, vehicles, managed sampling. In his photos, the cut is not dramatic. The consequence is. At the wound margin, cell behavior shifts. He annotates the image with notes about fibroblast recruitment and chondrogenic fronts, about a callus that forms quickly and refuses the modesty we associate with repair (Price et al., Deer Antler Regeneration, 2005). In the field he will summarize it more plainly. Repair here doubles as reorganization. I write that down and then wonder, is the callus closing a gap or proposing a new plan.

The antler continues in that altered state. Mark measures deviation as angle and curvature, as thickness that persists where taper used to be. He will say that signaling gradients have been deflected. The map changed locally, so trajectories reroute. I translate that for myself as a structural accent that begins to look like a decision. By mid summer, velvet dries and peels, and the antler mineralizes, locking in the season’s edits (Bubenik, The Physiology of Antler Growth, 1990). He calls it a record. I look at it like a paragraph where a sentence was crossed out and rewritten just enough to be legible.

When autumn gives way, the antlers are cast. Mark is precise here. Osteoclast activity defines a separation zone at the pedicle junction. The fall is the endpoint of a cycle, not an erasure (Lincoln, Seasonal Cycles in Deer, 1998). He collects sheds and tags them by year and location. In his notes, the word reset appears rarely. He prefers carryover. I take that word and try to test it against what I see.

Because the pedicle remains, and with it a starting condition that is no longer neutral. Mark’s longitudinal sets matter here. The following spring, new antlers rise from the same bases, and he looks for echoes. Not identical features, but constrained deviations that recur in place. A thickened ridge where last year’s callus sat. An outward bias in a tine that should have tracked straight. He is careful with language. He does not call this inheritance. He calls it persistence of altered state. I hear the caution and still feel the pull of the other word. When a pattern comes back, what else are we supposed to think.

Mark’s answer lives at the level of systems. Developmental programs reengage each year. The familiar genes are involved, HOX clusters, FGFs, BMPs, WNT pathways, establishing axes, branch points, growth rates. None of that surprises him. What matters is how those programs start. Cells that endure in the pedicle carry forward regulatory configurations from the previous season. Their chromatin landscapes, their methylation patterns, the simple fact of which loci are accessible, bias the next round of growth without altering the DNA sequence itself (Kierdorf et al., Frontiers in Bioscience, 2009; Price et al., Journal of Experimental Zoology, 2005). He shows me side by side images and asks a question he already knows I find uncomfortable. If the same script reads differently because the margins changed, where do you locate the change.

I try to answer without collapsing categories. Within a single animal, the body can remember by persisting in state. Across generations, that memory mostly resets when germ cells are formed and regulatory marks are reprogrammed. Mark insists on this boundary in his papers and in conversation. He has a way of drawing two timelines in the air with his hands, then refusing to let them overlap. I accept the diagram, then return to the field and feel the overlap anyway. The repeated morphology creates a convincing illusion. It looks like heredity because it recurs, because it constrains variation into familiar shapes, because our eyes reward stability.

We test that illusion against a simple case. A buck with a velvet injury near a branch point. Mark documents it. Next season, an altered tine appears again, not identical but consistent. He writes that the underlying genome has not changed with respect to germline transmission. I write that the body edited itself and then replayed the edit. Both statements are true, and one of them keeps misleading people.

So here is the dilemma that follows me out of Mark’s datasets and back onto wet ground. If repetition can arise from carryover of cellular state, how often do we mislabel it as inheritance. When you see a feature persist, what evidence would you demand before you claim it belongs to DNA rather than to the way DNA is used. Mark would say the burden is high, and that most antler variations after injury fail that test. I agree with him when I am reading. I hesitate when I am watching a stag lift its head and the same irregularity returns.

Mark’s discipline is to separate persistence from transmission. Mine, if I can call it that, is to notice how easily the former impersonates the latter. Between those positions, the deer stands each spring with antlers that are both familiar and revised, offering a structure that remembers without reproducing its memory in offspring. The illusion holds just long enough to teach you something, and then it breaks if you insist on following the cells.

Morphogenesis and Regeneration: What Is Actually Happening

Mark handles the pedicle the way some people handle instruments, with a kind of quiet certainty about what it will do if left intact. He will tap the base of a shed and say, this is not just a stump. It is an active field, vascular, innervated, and primed to restart a developmental program that looks uncannily like embryology rerun in an adult body (Goss, Deer Antlers: Regeneration, 1983). I used to think of antlers as seasonal add‑ons. After following his data, I do not. The pedicle behaves less like a foundation and more like a persistent source of positional information. It sets the coordinates before anything grows.

That starting condition matters because the system does not build from scratch. It replays. When spring physiology shifts, stem‑like mesenchymal cells within and around the pedicle proliferate under the velvet. Mark tracks this phase closely. He samples growth rates, vascular density, local gradients of signaling molecules. He insists on thinking in gradients rather than parts. Cells respond to concentrations of factors rather than to preformed instructions, he says, and that difference changes how you interpret form.

The signaling network he points to is familiar but not trivial. HOX genes establish positional identity along the proximal to distal axis, giving the antler a sense of organized direction (Kierdorf et al., Frontiers in Bioscience, 2009). FGFs push proliferation and elongation, maintaining the outgrowth zone where new tissue is continuously added (Price et al., Journal of Experimental Zoology, 2005). BMPs regulate ossification and branching decisions, interacting with local mechanical forces and tissue density (Bubenik, The Physiology of Antler Growth, 1990). WNT signaling coordinates cell fate and pattern stability, helping maintain the coherence of the structure as it extends (Kierdorf and Kierdorf, Antler Regeneration, 2011). Mark says these pathways are reused rather than invented. The antler is not special because of unique genes. It is special because of how ordinary developmental programs are reactivated with unusual intensity.

I watch him map these interactions, and I notice something else. He is always asking about initial conditions. What does a cell think it is when growth begins. That question shifts everything. If HOX expression domains are slightly displaced, if FGF gradients are skewed, if BMP signaling is locally elevated or suppressed, the resulting morphology diverges without any change to DNA sequence. The system is sensitive in the way weather is sensitive. Small differences early on amplify into visible structure.

This is where I bring in Kris, my niece, geneticist… more about her later.

Now bring injury into that system. Mark does not treat it as an external interruption. He treats it as a reconfiguration of the field. When velvet is cut, the wound site becomes a competing center of signaling. Inflammatory cues overlap with developmental pathways. Growth factors surge, progenitor cells flood in, and the boundary between repair and development dissolves almost immediately (Price et al., Deer Antler Regeneration, 2005). I used to think repair would simply restore what was lost. It does not. It introduces a new set of coordinates.

Is repair ever just repair in a system that grows this way. Or is it always a proposal for an alternative structure.

The callus that forms at the injury site is the clearest example. Mark describes it as exuberant, which sounds subjective until you see it quantified in thickness and growth rate. It behaves like developing antler tissue, not like a cautious patch. BMP activity ramps in ways that promote ossification but also alter branching thresholds. FGF signaling sustains local proliferation longer than it would have without injury. WNT pathways stabilize the new configuration so it does not collapse back into the old pattern. What emerges is not a corrected antler. It is an altered trajectory that looks consistent enough to be mistaken for intention.

I carry that observation back into what I see from the field. A bend where straightness used to be. A new tine that appears as if it were always planned. The next year, something like that bend returns. Not identical, but constrained. Mark calls this iterative morphogenesis. Each cycle begins with altered inputs derived from the previous one. Cells that persisted in the pedicle and adjacent tissues retain regulatory states that bias gene expression when growth resumes. DNA sequence has not shifted. The deployment of that sequence has.

We run into the same pattern any time someone brings up muscle or skin. You lift for a few months and your strength increases, fibers thicken, coordination shifts, and if you return after a break the gains come back faster than they did the first time. People call it “muscle memory” without much hesitation. The phrase sticks because something clearly persists. Satellite cells retain altered states, transcriptional responsiveness shifts, and prior training changes how quickly the system ramps back up (Seaborne et al., Human Skeletal Muscle Epigenome, 2018). Or take something even simpler. Spend a few summers outside and your skin darkens more quickly each year. Melanocytes respond differently after repeated exposure. No one argues that your DNA has changed each time. They still treat the responsiveness as a kind of inheritance, just compressed into a single lifetime. The logic is identical to what we see in antlers. A system absorbs experience, the next iteration runs under modified conditions, and the outcome looks familiar enough to feel encoded. The difference is not in the mechanism. The difference is that outside of antlers, people are more comfortable leaving it unnamed.

This is where I see people begin to misname what they are looking at. If a structural feature recurs, the word inheritance appears almost automatically. Mark resists it. He says persistence of state. He points to epigenetic configurations, methylation patterns, chromatin accessibility, spatial architecture in the tissue that constrains diffusion and signaling (Fraga et al., Epigenetics and Regeneration, 2005). These are not abstract details. They are the reason the same program produces different outputs across years within the same animal.

So what, then, is repeating. Not a blueprint in sequence, but a bias in execution. The pedicle carries forward a set of altered starting conditions. When HOX domains reactivate, they do so in a field that has been slightly shifted. When FGF gradients establish, they flow through tissue that has already been reorganized. When BMP thresholds are reached, they are reached sooner in some regions because density and composition changed. WNT signaling stabilizes whatever configuration emerges, even if that configuration originated in injury.

I find myself asking a question that I did not expect to matter this much. If the same pathways are used every year, why do they not converge back to the original shape. Mark answers it without hesitation. Because the system is path dependent. Once you alter the conditions that shape gradient formation, you alter all subsequent outcomes. There is no requirement to return. There is only a requirement to remain coherent.

Consider a real case Mark showed me, two consecutive seasons from the same individual. First year, a mid‑shaft injury produces a bulbous callus and a slight fork where none existed. Second year, no injury. Yet the fork recurs in attenuated form, and the shaft thickens earlier than expected. The pathways involved are the same. The starting state is not. That is enough.

You could frame this as a philosophical problem, but it is more uncomfortable than that. It asks whether we trust sequence too much as the sole carrier of biological information. Within an organism, clearly we should not. The antler demonstrates that developmental systems store information in configuration, not just code. But here is the dilemma that Mark keeps forcing onto the table. Does that configuration cross into the next generation. In almost all cases, no. Germline resetting strips most epigenetic marks, and offspring begin with a different baseline. The recurrence you see belongs to the individual, not the lineage.

There is an older idea hiding just under that instinct. Lamarck’s version of evolution never fully disappeared because it maps cleanly onto what people think they see. Use changes form, those changes persist, persistence becomes inheritance. It fits experience too neatly to let go without resistance. When people look at something like the antler and feel that pull, they are not inventing a new mistake. They are drifting back toward a familiar one, where lived change is assumed to cross the boundary into lineage simply because it endures long enough to be noticed.

This is why it looks hereditary. Because iteration produces familiarity. Because constrained variation produces patterns our minds categorize as traits. Because we are primed to see repetition as lineage rather than as memory embedded in tissue.

Mark stays disciplined. He measures, annotates, separates timelines. I absorb his work and still feel the pull of the illusion when I am back in the field. The antlers rise again, altered in ways that echo prior seasons, and the structure argues for continuity in a voice that sounds like inheritance. The mechanisms argue otherwise. HOX, FGF, BMP, WNT, reactivated and redirected. Repair programs and developmental programs interwoven until they are indistinguishable in outcome.

If you had to decide, standing there, what you are looking at, what would you call it. A genetic trait reasserting itself, or a history that has been carried forward and replayed without crossing into reproduction. How much repetition do you require before you assign it to DNA. How much deviation convinces you it is not.

Mark would say the answer is in the mechanism. I am less certain. I think the answer is also in how easily our perception collapses different kinds of memory into one. The deer keeps growing structures that feel inherited. The biology keeps demonstrating that they are not. Between those positions sits the actual process, iterative, state dependent, and stubbornly resistant to simple labels.

Next in Part 2 Epigenetics: What It Actually Is

References

Allis, C. D. et al. Epigenetics. Cold Spring Harbor Laboratory Press, 2007.
Bird, A. “DNA Methylation Patterns and Epigenetic Memory.” Genes & Development, 2002.
Blewitt, M. E. et al. “Dynamic Reprogramming of DNA Methylation at an Epigenetically Sensitive Allele in Mice.” PLoS Genetics, 2006.
Brockes, Jeremy, and Kumar, Anoop. Appendage Regeneration in Vertebrates. 2005.
Bubenik, G. A. “The Physiology of Antler Growth.” Journal of Experimental Zoology, 1990.
Buenrostro, J. D. et al. “ATAC-seq: A Method for Assaying Chromatin Accessibility.” Nature Methods, 2013.
Candia Carnevali, Maria. Regeneration in Echinoderms. 2006.
Feil, R., Fraga, M. F. “Epigenetics and the Environment: Emerging Patterns and Implications.” Nature Reviews Genetics, 2012.
Fraga, M. F. et al. “Epigenetic Differences Arise During Tissue Regeneration.” Nature Reviews Genetics, 2005.
Futuyma, D. J. Evolution. Sinauer Associates, 2013.
Goss, R. J. Deer Antlers: Regeneration, Function and Evolution. Academic Press, 1983.
Graf, T., Enver, T. “Forcing Cells to Change Lineages.” Nature, 2009.
Gurtner, Geoffrey et al. Wound Repair and Regeneration. 2008.
Heard, E., Martienssen, R. A. “Transgenerational Epigenetic Inheritance: Myths and Mechanisms.” Cell, 2014.
Heijmans, B. T. et al. “Persistent Epigenetic Differences Associated with Prenatal Exposure to Famine.” Proceedings of the National Academy of Sciences, 2008.
Jablonka, Eva, and Lamb, Marion. Evolution in Four Dimensions. 2005.
Jaenisch, R., Bird, A. “Epigenetic Regulation of Gene Expression.” Nature Genetics, 2003.
Kahneman, Daniel. Thinking, Fast and Slow. 2011.
Kierdorf, U. et al. “Antler Regeneration and Its Underlying Mechanisms.” Frontiers in Bioscience, 2009.
Kierdorf, U., Kierdorf, H. Antler Regeneration in Cervids. Springer, 2011.
Kouzarides, T. “Chromatin Modifications and Their Function.” Cell, 2007.
Lincoln, G. A. Seasonal Cycles in Deer. Cambridge University Press, 1998.
Mayr, Ernst. What Evolution Is. 2001.
Meaney, M. J. “Epigenetics and the Biological Definition of Gene–Environment Interactions.” Child Development, 2010.
Morgan, H. D. et al. “Epigenetic Inheritance at the Agouti Locus in the Mouse.” Nature Genetics, 1999.
Price, J. S., Allen, S., Faucheux, C. “Deer Antler Regeneration: A Stem Cell-Based Epimorphic Process.” Journal of Experimental Zoology, 2005.
Reik, W. “Stability and Flexibility of Epigenetic Gene Regulation in Mammalian Development.” Nature, 2007.
Sánchez Alvarado, Alejandro. Regeneration in Planarians. 2006.
Tanaka, Elly. Regeneration in Amphibians. 2016.
West-Eberhard, M. J. Developmental Plasticity and Evolution. Oxford University Press, 2003.