Information about activities. How to join the TT. Discussion list, for members and others! Species-specific Care sheets Articles from our Newsletter. Links to important sites. Links to members' sites.Search our site! What has been updated, and when.

The causes of “Pyramiding” deformity in tortoises: a summary of a lecture given to the Sociedad Herpetologica Velenciana Congreso Tortugas on October 30 2010

A. C . Highfield - Tortoise Trust

English text first published online Nov 1 2010.

What follows is a brief outline of the major findings of our work in establishing the precise physical and biological mechanisms involved in “Pyramid Growth Syndrome”. This has been a very complex and demanding piece of research. This is a problem we have addressed continually since around 1990, but from around 2004 we intensified efforts to find some answers. In the course of this, we have conducted extensive fieldwork, have used diagnostic image techniques, and have conducted multiple post-mortem and laboratory examinations of both normal and effected animals. I would stress that at no time was any animal killed or hurt in this process. We relied exclusively upon ‘natural’ casualties from other causes. Key objectives were to look at competing theories, to separate those that had some factual basis from those based upon incomplete or false data, and to conclusively establish the exact mechanisms involved in producing the effect. A further objective was to begin to develop some practical guidelines whereby the problem could be alleviated or prevented.

This particular article is merely an extract from the complete and comprehensive review we have prepared (1). That will be published separately in the near future. It contains extensive data and is fully referenced. The purpose of this article is merely to set out the basics in a form that is easy to access and is easy for typical keepers to understand.

Two key theories have dominated discussions on this subject:

The abnormal growth is caused by incorrect diet, specifically by high protein, high energy and calcium deficient diets;

The abnormal growth is caused by lack of humidity or by general dehydration or both.

As we shall see, all of these factors play a critical role.

One of the major problems with what we will call “the humidity theory” has been a lack of any biologically credible explanation for it. Some suggestions have been made, but these require us to abandon established physiological science, and grasp at vague concepts of “cellular dehydration” and “tissue collapse”. Other theories along the same lines have proved similarly unviable. However, it was clear that many keepers were observing some kind of effect linked to humidity and also to generalised levels of hydration and heat. One of our most important objectives was to try to understand what was really happening.

The first thing to point out is that chelonia are constructed from exactly the same materials as many other animals. Their skeleton, though different in form, is materially virtually identical to that of other species. In a similar manner, the outer keratin scutes are comprised of primarily beta-keratin with some alpha-keratin cells also present. These are well studied materials. What is unique in tortoises (and turtles) is the way in which the skeleton surrounds the body, and the extensive covering of the keratin layer. Consequently, any disruption of either of these layers will have a profound effect.

If we first examine the bony skeleton, we note that this is vulnerable to exactly the same diseases of deficiency as that of a dog, a horse or a human being. There is absolutely nothing unusual or unique in how a tortoise skeleton develops and is sustained. The process is entirely normal and is completely consistent with established biological and nutritional knowledge.

To develop normally, the skeleton requires a supply (carried by the blood, and obtained from the food intake) of essential bone-building trace elements, principally calcium and phosphorus. These trace elements need to be in the correct quantities and proportions. In addition, in order to transport these materials, the animal’s vitamin-D3 metabolism must function correctly. Any failure, ether of the supply of “raw materials” or of the transport (D3) mechanism will result in bone formation that lacks normal density and strength. In reptiles, this condition is widely recognised by keepers as “MBD” or Metabolic Bone Disease.

Bone that lacks adequate density is highly vulnerable to physical stresses and is liable to deformity as a result of conflicting physical forces. This is perhaps best known in humans as the condition called “Rickets” where the long bones of the legs are bowed and bent by a combination of the effect of gravity (weight) and the pull of muscles. If we examine the structure of bones affected by Rickets, we find a remarkably similar condition to that seen in tortoises suffering from MBD and so-called “Pyramiding”. Instead of being hard, thin and dense, the bones are fibrous, thick and porous. Such bones are extremely easy to deform under conditions of constant stress. In a tortoise, one cause of stress is that of the attached (and very powerful) muscles of the limbs. It is not unusual to see tortoises that suffer from MBD or Pyramiding to also display a depressed pelvic region. The cause of that is muscular tension. Such animals may also display a ‘bulging’ effect at the upper body - caused by lung expansion and contraction and the muscles involved in that process. It is during phases of growth that bone is most vulnerable to such effects. It is at its most plastic. The more rapid the growth is, the greater the potential for absolute or relative deficiencies to occur. This is a well known relationship and affects all animals, and humans. It can be extremely challenging to obtain good bone density in captive situations on artificial diets. Herbivores are especially sensitive. On greatly accelerated growth regimes achieving normal, healthy bone density is extraordinarily difficult. In fact, so much so that to date I have never seen it. 100% of the animals we have examined that were raised on high growth regimes have MBD present to some extent, whether or not they displayed obvious external symptoms. This is relatively easy to demonstrate by dissecting deceased specimens, or by comparing x-rays of wild tortoises to captive raised examples.

Every tortoise breeder will know that as a hatchling emerges, the bones are extremely soft and flexible. They gradually harden up over the following days, months and years. However, they are never completely fixed. They continue to remain sensitive to stresses applied over an extended period. Even relatively slight tensions, applied continually, can have a substantial effect.

A key area where tortoises and most turtles differ from other animals is, as we have noted, the fact that they are largely surrounded by their skeletal structure. The bony, internal layer is overlaid by keratin shields or scutes. Keratin is an interesting material with many unusual properties. It is one of the strongest and toughest biological materials known to science. It is also hygroscopic, and loses and absorbs moisture in an effort to reach equilibrium with the environment. This effect is known to everyone: after a hot bath or shower our fingernails are soft and pliable. After a day hiking in the desert, they are hard and brittle.

Both Alpha and Beta keratins have been studied extensively, and we know quite a lot about how they perform at differing levels of moisture content and in different levels of ambient humidity. One very important feature is the way in which they gain and lose stiffness as they respond to external humidity. These changes are dramatic. They can be measured and quantified. At levels of humidity above 80% scute keratin possess only a fraction of the strength and durability that it does at 50% RH. Such changes can be measured and quantified using criteria such as Young’s Modulus. At sustained levels of 90-100% it is essentially saturated as it accumulates water molecules rapidly. It becomes extremely soft and pliable, exerting almost no stress on the underlying skeleton. At the opposite extreme of low humidity, below approximately 25% it loses water molecules and becomes very stiff and resistant. At such levels it is exerting a very substantial physical force on the bone beneath. We know from earlier tests we conducted with tortoise vivarium design, that many of these create extremely dry conditions, with sustained relative humidities as low as 12%. More recent tests have also shown that directly under basking (heat) lamps very low, very localised conditions of humidity well below 20% can occur immediately adjacent to the surface of the scutes of the carapace. This has a profound drying effect, increasing keratin stiffness, driving out water molecules, and at the same time increasing stress forces upon the underlying bony skeleton.

It is critically important to address one very common piece of misinformation in this context. It has been claimed that juvenile tortoises (for example, Testudo graeca) spend most of their time in the wild in “humid” microclimates where ambient conditions are in the range of 90-100% RH. This is completely false. One part of our study involved taking many thousands of measurements in the natural habitat to establish the actual conditions experienced. Our methodology involved the use of miniature automatic data loggers that recorded both temperature and humidity with a very high degree of precision. We took recordings over a complete 12-month cycle in several key habitats. We also attached loggers to tortoises and recovered them later to collect the data. In total, we collected 18,000 data points detailing humidity alone. What we found - in brief - was that juvenile tortoises were not experiencing substantially different levels of humidity than adults. While it is perfectly true that tortoises make extensive use of selected microclimates, the levels recorded in these were in the range of 34-60% RH. The sole occasions when levels in excess of 90% were recorded were during thunderstorms and episodes of rain. In the semi-arid environments of Ameria and Murcia (which are very similar to those found in most of North Africa) rainfall is not a regular occurrence, even during peak periods of tortoise activity. In total, we found that tortoises were only exposed to levels of humidity that could reasonably be described at “high” (80%>) for 2% of the total time recorded. While foraging humidity could be as low as 20%, but this was followed by retreat into vegetative microclimates where typical levels average around 45-50%. Measurements were taken from tortoises in all activity phases, include those buried in scrapes. This data accords with that previously recorded (though not in such detail) in Morocco, Turkey and Tunisia. Other workers have also conducted extensive humidity data collection from tortoise burrows of the Desert tortoise, in North America. This data also fails to reveal any evidence whatsoever of the availability of “humid” hides offering relative humidities of “90-100%”. In fact, levels in the arid desert regions of Arizona occupied by Gopherus agasssizii are even lower than those we recorded here in Almeria and Murcia. If indeed (as claimed) the main reason why wild tortoises in arid regions do not suffer from “pyramiding” is because they make extensive use of “caves” or “vegetation” offering “90-100% RH” retreats then all of our local Testudo graeca graeca in Spain would be very lumpy and deformed indeed, because no such facilities exist. This is a semi-arid arid habitat with some of the lowest annual rainfall in Europe, with an average precipitation of just 226mm per year (the UK receives almost 600mm).

I would urge all keepers to exercise extreme caution in relying upon generalised “average” figures for relative humidity in tortoise habitats obtained from climate sites. These can create a very misleading impression indeed. The only meaningful, reliable data is that collected at the level occupied by tortoises (not several metres above ground) and in the precise microclimates used. There are massive differences in temperature and humidity at various altitudes, and also conditions inland can vary substantially from those closer to the coast. Tortoises tend to occupy very narrow, very specific biotypes. This needs to be understood when considering climatic data. Data based on an “average” of all habitats within a region or country is likely to be completely misleading and inappropriate.

There is a further important feature of keratin that affects physical stress on the skeleton and the modes by which it proliferates in chelonia. There are two primary methods. Tortoises exclusively rely upon a mode of cell proliferation that deposits new material at the edges of the scute, resulting in the well-known “tree ring” effect. Aquatic turtles mostly (not all) rely upon a mode where new growth occurs in a horizontal plane only, with new cells growing at an even rate beneath the older material. Many species of aquatic turtle lack “historical growth rings” for this reason. Eventually, the older scute is shed (usually entirely) to be replaced with the new, larger scute. In terrestrial tortoises, this pattern of shedding does not occur. The keratin builds up, in a vertical mode, continually.

This vertical and expansive mode of cell proliferation of itself creates an upwards force on the skeleton. Where there is concurrent MBD of any degree, the effect will be substantial. The bone will attempt to conform to the pattern of growth of the scute. This is the primary mechanism involved in “pyramid” growth syndrome in tortoises and the main reason why it is not observed in those aquatic species that shed whole scutes. It is further amplified in effect where certain conditions apply:

Where the keratin is excessively stiff as a result of very low humidity.

Where the keratin is unnaturally thickened.

Thickened keratin, with over-proliferation, is a common feature of every severe case of “pyramiding” observed to date. It can often be readily diagnosed by comparing the colour of the keratin to a wild, healthy specimen. In cases of over-proliferation, it is typically much denser and much darker. These animals tend to share some history in common. In the majority of cases, this includes being raised in vivarium systems, under heat lamps, and frequently deprived of adequate fluid intake. We have already noted that many vivarium systems produce incredibly low, sustained levels of ambient humidity, far below that experienced even by tortoises from the most arid habitats. The effect of an overhead heat lamp on a tortoise has been inadequately studied, but it certainly has a very powerful dehydrating influence. Tortoises kept in such conditions also tend to present with other health issues related to generalised dehydration: bladder stones, gout, and renal failure. All of these are associated with sub-optimal hydration.

One very interesting effect has been demonstrated in laboratory tests with chelonia. As they are subjected to extended periods of dehydration, the epidermis thickens in an attempt to reduce cutaneous evaporative losses. This affects the skin of the limbs, and in particular the proliferation of beta-keratin that comprises the horny scutes. As the animal is subjected to dehydration, the scute growth accelerates, becoming ever thicker. Bone growth however does not accelerate at the same rate, producing a major differential. This thickened, dry keratin begins to exert an enormously amplified force upon the skeleton (which in such animals is typically of very poor density). This is one other very important reason why we tend to see particularly badly deformed animals that have been raised in conditions of sub-optimal humidity. Where accelerated growth (and typically MBD) meets dangerously low levels of humidity the conditions are ideal for producing gross distortions of the carapace due to the conflicting physical stresses of muscular tension and tension resulting from over-proliferation of the keratin scutes.

A further important factor is that wild terrestrial tortoises are continually wearing and abrading excess keratin from the scutes in the course of normal life. Unlike turtles, they do not shed old keratin - they wear it away. They are abraded by coarse vegetation, by impact with rocks, by wind-born sand particles, and by the constant burying, digging and excavating they engage in. While estivating or hibernating they are not motionless. They move, surrounded by highly abrasive particles in the soil. Also, micro-organisms in the soil gradually degrade the outer surface of the scutes. As such, they are subject to continual wearing (and resulting thinning) of the scutes. In the vast majority of captive situations (and especially so in indoor maintenance) this factor is totally overlooked by keepers. One result is that the keratin continues to accumulate even if humidity is not an issue. Where very low humidity is also present, the effect is compounded.

The thicker (and dryer) the keratin, the worse the “pyramiding” becomes. This is not only a consequence of the keratin layer itself, but also of its powerful physical deforming effect on the skeleton immediately beneath.

Some keepers have mitigated these obvious symptoms by essentially “soaking” the animal at high ambient humidity (in the 90%+ range) and at elevated temperatures for long periods. The real effect of this is simply to soften the keratin and reduce stresses. This technique does nothing to address any underlying MBD (it just makes it much less obvious) and in addition it exposes the animal to a high risk of fungal and bacterial infections. The soft, wet keratin is highly vulnerable to such organisms. It lacks structural integrity and damages easily. While the visible “pyramiding” may be reduced or eliminated by such extreme measures, this is not a very satisfactory answer. It is an artificial solution to an an entirely artificial problem.

If a proposed husbandry method requires us to resort to totally unnatural conditions (extended exposure to 90%+ RH for arid habitat tortoises) to solve a problem created by other totally unnatural conditions (sub-optimal humidity, dehydration and excess growth rates resulting in poor bone density) then I would suggest that something is seriously wrong. It is far preferable to address the fundamental causal issues and not to merely concentrate on trying suppress the most obvious symptoms, which is the sole result of such methods.

In this case, the issues that need to be addressed are:

Obtaining growth that results in comparable bone density to healthy wild examples. This should be tested by routine radiography. By obtaining healthy density maximum resistance to subsequent deformity is achieved.

To ensure that captive environments provide a suitable range of both humidity and temperatures. A safe and appropriate range should be established by reference to reliable data from the natural habitat of the species in question, not by guesswork or by reference to inappropriate and frequently misleading general climatic averages.

Looking again at captive enclosures to develop methods of allowing natural wear and tear of carapace keratin to occur, thus avoiding a continual build-up.

There are challenges in meeting these needs. It is very clear that the design of captive habitats needs to improve. Heat lamps do indeed assist in some ways, but from this study it is also clear they can have a very serious negative effect. Obtaining safe levels of ambient humidity is also not easy for keepers outside the natural range, who must rely on artificial enclosures. Obtaining high-grade bone density and sustainable rates of healthy growth are also very difficult in captive situations. Wild tortoises practice cyclic activity and feeding, with long periods of non-feeding and inactivity. The very nature of their intake also varies seasonally, especially in terms of moisture and fibre content. Practical solutions to these problems are not beyond the bounds of possibility however, and will represent a major step forward in chelonian husbandry. This is especially important in conservation breeding for eventual release or when propagating highly endangered species. It is extremely important that compromised animals must be avoided in such situations.


I would like to express my thanks to Ed Pirog for many fascinating debates, discussions and indeed arguments on this topic over almost two decades. It is true to say that Ed has been as persistent as I have in trying to understand this problem. Ed had made some important observations on this issue, and also identified the keratin layer and both humidity, heat and hydration as playing an important role in a problem that has plagued keepers ever since captive breeding of tortoises became commonplace. It was, in part, due to Ed’s observations that I set out to examine the role and performance of keratin in different environmental conditions in as much detail as possible and it was these investigations that ultimately led to the results presented here for the first time.

(1) Humidity, Growth, Physical Stresses and the Development of Carapace Abnormalities (‘Pyramiding’) in Tortoises: A review of current research (in preparation).

(c) A. C. Highfield / Tortoise Trust 2010