6. The growth of heteromorph shells

Heteromorph ammonites, forms which have given up the normal spiral, have all along posed the greatest problems of understanding upon palaeontologists. It was early recognized that these forms could hardly have been suited for a swimming lifestyle. Many workers were inclined to regard them as strange bottom-dwellers because of a frequently missing bilateral symmetry. Since heteromorphs were particularly abundant during the Cretaceous and since all ammonites disappeared at the end of this period, these forms were seen as an expression of the general ammonitic decline, degenerated final members. However, this view has been changed by the work of J. Wiedmann (1969) and G. Dietl (1978). Nowadays they represent entirely vital, though specialized forms of the ammonite spectrum. However, I have certain doubts whether they were really specialists. It is only the shell shape that differs from normally coiled ammonites. This fact may have been of little importance for the living individual.



Fig.17. Various heteromorph forms which give an impression of the enormous variability 


Heteromorph ammonites with similar shapes as those from the Cretaceous appeared already in the Triassic as well as in the Middle Jurassic, although these forms were not very frequent. Despite the problems which a swimming heteromorph would have had the idea that these forms were swimmers has never completely been abandoned. If they were unable to swim actively they should have been passive floaters in the blue seas. This may appear plausible for certain adult shells with a final hook (fig.17), but during growth they would have had great difficulties as to their apertural orientation due to frequent changes (fig.18). An advantage of a frequently changing orientation in the water is hardly recognizable and certainly disadvantageous for nutrition



Fig.18. Balanced shell positions during growth stages in some heteromorphs, swimming   ability supposed


Not all heteromorphs differ completely from normally coiled forms. In many cases the deviation from the normal spiral is marginal. It can be restricted to a local loosening of whorls which later become tight again. Many are uncoiled only in early stages and return to the normal spiral, others are normal in the beginning and become uncoiled in later stages. Such variations can serve as an indication that the mode of life in heteromorphs differed hardly or not at all from normally coiled ammonites.

 The strange heteromorph forms offer many interesting informations about their individual growth history, in any case much better ones than could ever be derived from normally coiled ammonites. It is very advantageous that even the knowledge of soft parts protruding from the shell ist not at all required. Only the shell growth must be followed up step by step, taking account of the prevailing physical conditions, that is the differing ratio of weight to buoyancy. The behaviour of the first shell part moves the points for the following shapes of the growing shell. Obviously, the animal was unable to take an influence over the shell shape and had to accept it for better or for worse. 

Since I could prove the missing ability to swim already for normal ammonites this result should also be valid for heteromorphs. This can, fortunately, be shown using a different approach which is not based on calculations.

Meanwhile it has become clear that things have been quite different from former suppositions of palaeontologists.


6.1. A simple experiment explains the basic conditions in heteromorph shells

Heteromorph ammonites are particularly interesting because they allow observations which are impossible in normally coiled forms. When I started dealing with the lifestyle of ammonites I knew from the very beginning that a relevant model had to include the heteromorphs. They were an important key. I had to test a new idea in any case in heteromorph shells. Only if they did not reveal a conflict the idea could be correct.

As a particularly interesting feature I found that many orthocones possess an oblique sculpture ( figs.17, 18). Such shell parts occur exclusively before the formation of the first U-shaped hook, and in each case the sculpture shows the same inclination with respect to the following shell parts. This frequently recurring sculpture has occupied my mind for a long time, because it did not appear accidental to me, rather happened in a regular way and during certain ontogenetic stages. Finally, the idea struck me that this fact is connected with a definite orientation of the shell during growth. In addition, in such forms the sutures are very simple. After finally calculations had confirmed the benthic lifestyle I could find the solution of the problem with the oblique ribs.      

The decisive questions are: 1. which positions can an orthocone shell assume in water?

                                                2. is there an unequivocal solution which is in harmony with real shell remainders ?

Just like a normally coiled ammonite an orthocone shell has a gas-filled phragmocone and a body chamber containing soft parts. This arrangement means that centre of gravity and centre of buoyancy have different positions, depending on the weight distribution. While the centre of buoyancy as the centre of volume has an invariable position this condition does not apply to the centre of gravity. The longer the body chamber is the shorter is the distance between these centres.



Fig.19. The four basically possible stages of balance in a test-tube filled with water to a variable extent. A small additional weight on the bottom of the tube simulates the soft body and facilitates the establishment of the oblique position. The weight increases from left to right.


I have simulated the orthocone shell using test-tubes filled to a different extent with water. The low weight of the cork was adjusted by a small additional weight. At the bottom another small weight was attached to simulate the higher weight of the soft body compared to water. This experiment can easily be repeated by everybody. It shows that there are exactly four positions which an orthocone shell can assume in water. Three of these positions are self-explanatory (a, b and d in fig.19)


In fig.19a the shell is lighter than water and ascends to the surface in a vertical position. In fig.19b the shell is slightly heavier than water and rests on the ground in a vertical position as well. Between these two positions is the stage of neutral buoyancy. It is very sensible, and I had no success in establishing it experimentally. It should be noted that all of these three stages are restricted to a vertical shell position. This is also the rest position in the Recent Spirula which has a phragmocone at the rear end.

Fig.19 a and b make clear that obviously the task of the phragmocone did not consist in establishing neutral buoyancy but presumably it had the different purpose to establish a stable more or less upright position in order to help carry the shell easily. This is a very important task which should not be underestimated. In addition, of course, the phragmocone serves the purpose to protect the soft body at its rear end after the advancement to a new position within the shell.

In the position as in fig. 19d the test-tube is almost completely filled with water and horizontally lying on the ground. The  position lying on the ground is the only horizontal one that an orthocone shell can assume.

Neutral buoyancy and simultaneously a horizontal or inclined position are unfortunately absolutely impossible!

 Such an idea is based on the fantasy of dreaming laymen, far away from reality.

Indeed, there are many early orthocone nautiloids, for example from Gotland, in which the chambers of the phragmocone are filled to a large extent with calcareous deposits which indicate that these shells really did have a horizontal orientation. This can only mean that these shells already during life were lying on the ground. Maybe, these animals filtered their food from streaming water. The sediments of Gotland indicate shallow agitated water. The strong weight increase of the filled phragmocone strengthened the position on the ground and helped to prevent a.displacement by currents. Between the positions of fig. 19b and d we find a position in which the shell assumes an inclined one. This position is the most important one for understanding the formation of many heteromorph shells. However, it is valid for a wide range of inclination angles, the entire range between the upright and the horizontal positions. In fig.20, I present a somewhat more detailed sketch.





Fig. 20.  Stable  position of the shell with inclined orientation, acting forces in balance





This orientation is completely stable, that is, after a dislocation by moving water for example it returns automatically to its former position. This stable orientation is caused by the different positions of the centres of buoyancy and gravity. The moment of the weight (weight times lever-arm) which makes the test-tube descend to the ground is balanced by the erecting moment of the buoyancy. This function appears to be the main one of the gas-filled phragmocone and the original one, leading to a stable orientation of the protective shell,  if any further sense at all canbe provided by the shell. It should be noted that this condition is valid for a wide range of inclination angles, between rather flat and steep. This function may already have been active in the earliest forms such as Plectronoceras, even if the chambers in this form were stiill filled with water instead of gas, since water is lighter than the soft body and, therefore, leads to slightly different positions of the centres of gravity and buoyancy.

                                           6.2. Formation of heteromorph shells

In normally coiled ammonites in the region of the aperture the preceding whorl is surrounded by parts of the soft body, Thus, the shell growth in the symmetry-plane is automatically assured. However, in heteromorph ammonites the unrolling leads to a complete loss of control of the shell shape. Therefore, by chance manifold forms can develop. 

The oblique shell orientation in fig. 20 establishes the connection to orthocone heteromorphs, because it delivers an easy explanation for the oblique ribs and sculpture of many heteromorphs starting with an orthocone shaft preceding the first U-shaped hook. In this stage the weight is unequivocally higher than the buoyancy. However, it appears to be the lowest weight at all occurring in an ammonite. This is a curious result, since heteromorphs were the only ammonitic forms which some workers were inclined to regard as benthic. Obviously, this condition of low weight can easily be changed, visible by an increase in curvature. This causes problems to the animal which now is forced to balance the shell position by muscular activity. On the other hand, there is a rare case of reversed curvature (in Macroscaphites) which only occurs for a short time, indicating reduced weight. Presumably, weight changes occurred often in many ammonites, if not in all.

                                         6.2.1 Forms with an orthocone shaft following a loosely coiled initial shell

As a first example Ancyloceras is shown in fig. 21. This form is a succession of different weight stages. During the initial phase it is an almost normal ammonite with a somewhat reduced weight compared to nomally coiled forms. In the second phase (fig. 21b) the shell weight has decreased. Now the shell grows under self-balancing conditions. This conditions are maintained for rather a long time. But finally a stronger weight increase happens, and the animal is no longer capable of carrying its shell in an inclined orientation. It topples over (fig. 21c) and props on the bottom or on soft parts outside the shell. The overweight is rather small.



Fig.21. Growth stages in Ancyloceras with the shell position required and minimum soft body size outside the shell


In fig .21d the last growth stage of this individual is attained. Presumbly, it is difficult for many readers to get familiar with such an idea of ammonites, nevertheless, I cannot see an alternative. The soft body must definitly have been much larger than former ideas suggested. Soft parts protruded from the shell so that the shell could be carried during all growth stages without touching the ground.   

Only such forms which a final hook allow statements as to the nature of the soft body. Not only do they present informations about the minimum size outside the shell required, but also they give an impression about further details. The soft body cannot have been as compact as in a gastropod but there are indications that it possessed arms which allowed the shell to be held between them. Since the lateral symmetry was maintained in Ancyloceras the shell could hang down with the inial part below the aperture. This was the only way to carry the shell without damage. Also in normal ammonites indications can be found that the soft body was larger than presumed so far. For example the aperture in the tiny Oecoptychius refractus from the Middle Jurassic is so narrow that the soft body could hardly completely withdraw and had to stay outside. On the other hand the broad collars of certain lytoceratids point to a considerable widening of the soft body outside the shell. At least during secretion of a collar the mantle had to be connected to it. But unfortunately such features in normally coiled forms do not present a generally valid impression and might be interpreted as specializations of the soft body.

Contrary to former suppositions it is not the whole animal that in heteromorphs changes its position during growth, it is only the shell which follows the law of gravitation with the animal itself not being affected.

A similar course of growth as in Ancyloceras is found also in Macroscaphites shown in fig. 22, with the only difference that the first stage is entirely normal. Only in fig. 22b a decrease in weight happens which makes the shell reverse its spiralization. It appears that the weight reduction is stronger than required for the inclined position. For a short time the shell is higher erected. It is possible that such a low weight was disadvantageous, since in the stage as in fig. 22c the weight is slightly increased to conform with the self-balancing oblique orientation.



                                      Fig. 22. Subsequent growth stages in Macroscaphites


This position is maintained until the final hook is formed. During this stage the weight has become markedly higher to make the shell topple over and hang down from the aperture. Macroscaphites is one of few forms which I found with a reversed spiralization. Nevertheless, also in this stage the weight remained higher than the buoyancy. After the formation of a U-shaped hook a shell part with oblique sculpture can never again occur, at least not as long as a paralel shaft is present. The pendulous shell prevents it.  



                   Fig. 23. Subsequent growth stages in Pravitoceras


The development of Macroscaphites might easily have taken quite a different course. A shape as in the seemingly mysterious Pravitoceras shown in fig. 23 would have also been possible.

If the weight reduction continues a bit longer in the position fig. 23b, then a condition of indifference or even lability occurs, that is, in case of a following weight increase the shell can either assume a forward inclination or a backward inclination. At this time the animal has lost its control of the shell by soft parts. In Macroscaphites the shell has returned to the normal forward coiling direction, although an orthocone shell part is build. But in Pravitoceras a backward inclination happens. This leads to a change of the coiling direction, resulting in a reversed final hook. All these things happen accidentally.


6.2.2 Forms with an initial orthocone shaft

 The formation of a final hook is not imperatively required although frequently found. In many cases the whole shell consists of a straight shaft. Maybe such forms have died before the U-shaped-hook stage. This is the stage with the lowest weight and no pull being required. This condition results in an extremely  simplified suture. Under these conditions the soft body can slide down after formation of a new septum merely by its weight as it is the case in Nautilus. The attachment musculature only ensures a connection which prevents an



Fig. 24. Shell orientation in Bochianites and its very simple suture. It is a documentation of a missing pull on the one hand and on the other that the complexity of the suture is dependent on the size of the pull.


uncontrolled advancement. The simplified suture can only be found in straight or slightly curved forms. It does not occur exclusively in Cretaceous forms but also in Jurassic ones.

The inclined shell orientation of heteromorphs with an initial straight shaft has further consequences which are responsible for the shape of many very different forms.  

Since no muscular pull is required to keep the shell in its momentary position, that is no control is required, any orientation of the shell is stable, any deviation from the symmetry plane can happen without a problem for the animal. When somewhen during growth the shell weight increases and the shell topples over, a great variety of heteromorph shapes can develop.

Three different main types are possible which of course can be connected by intermediate forms :

           the shell topples over in the symmetry-plane, or

          the shell topples over under an angle of 90° with respect to the symmetry-plane, or

          the shell topples over under an angle of 45° with respect to the symmetry-plane.

Obviously, the initial orientation is accidental. When investigating heteromorph ammonites from the Middle Jurassic G. Dietl (1978) found helicoid forms with an approximately equal number of righthand or lefthand whorls. A corresponding result was found by U. Kaplan & F. Schmidt (1988) for Cretaceous hyphantocerids. There are even strong indications that the ammonitic shell shape is not at all genetically fixed, only the animal’s behaviour, which makes the animal secrete shell material regularly or advance in its shell and build a new septum and so on. In my paper on heteromorphs in 1992 I mentioned this idea and proposed to check it experimentally using planispiral gastropods. This proposal was adopted by A.Checa & A.Jiménez-Jiménez (1997). By the attachment of small additional weights to shells of Planorbis gastropods these authors could demonstrate that strong deformations occurred in all investigated shells, since the gastropods tried to carry the shell in the normal upright position. This experiment made clear that the shell shape is not genetically determined.

Nevertheless, all heteromorph shells are determined by universally valid physical laws.

        The initial shaft topples abruptly over and stays in the symmetry-plane (fig. 24)

Also forms such as Ancyloceras, Macroscaphites and Pravitoceras mentioned above belong to this group. According to the weight increase following the initial shaft forms consisting of several shafts can result, or in case of a low weight increase open spirals with many further possibilities. In forms such as Subptychoceras after formation of the initial shaft, which represents the automatical balance condition of the shell (fig. 24a), an abrupt weight increase occurs which makes the shell topple over (fig. 24b). Apparently, the soft body does not resist to the pressing shell. It becomes obvious that the soft body must be rather large so that the falling shaft does not touch the ground. This fact is indicative of the minimum size of the soft body outside the shell and is in sharp contrast to former suppositions. Moreover, there are very probably considerable differences concerning the whole organism between ammonites and the modern Nautilus. S. Rein (1997) has made probable great differences even between Nautilus and extinct nautiloids. Unfortunately, next to nothing is known about the soft body of ammonites.  



                                                                   Fig.25. Growth stages in Subptychoceras


The shell growth continues, and now the animal must move the growing shell upwards (fig. 25c). Since it is leaning against the previously built shaft, it cannot topple over again for some time. The oblique sculpture, therefore, cannot occur again during the remaining growth stages, because the shell can never again assume the oblique position. Apparently, the weight is so high that an automatical balance is not feasible.The animal now must actively balance its shell. The result is that a sculptural pattern develops as in normally coiled forms. Ribs run (always) essentially parallel to the ground. In fig.25d the shaft has grown so long that it must be balanced above the soft body which cannot be successful. Again the shell topples over. This process continues until the ontogeny ends.

The aperture has a downward orientation any time. Contrary to former ideas it is not the whole animal that changes its orientation during growth, but it is the shell which has to follow the laws of gravity. In forms with a final hook it finally hangs down from the soft body without touching the ground. The weight excess is considerable and presumably similar to normally coiled forms.

              The initial shaft topples slowly over in the symmetry-plane (fig. 26)



Fig. 26. Developing shell in Scalarites, starting with a straight shaft, and a resulting new shell shape as one possibility of several. After K. Tanabe et al. (1981), modified.


Under these conditions many further heteromorph forms can develop, with the hank-like Nipponites as an extreme possibility, if the initial whorl leaves the symmetry-plane to a variable extent. However, all of these forms follow also from the regularities given by physical laws. As opposed to the cases explained above in this case the weight increase at the end of the initial straight shaft is so low that the animal is capable of balancing its shell with the musculature available. Thus, an open spiral develops and results in forms such as Scalarites. Apparently, the muscular forces required are so low that the animal does not show a reaction when the shell moves out of the symmetry-plane to result in a moderately three-dimensional form.

   The initial shaft topples abruptly over and assumes a laterally inclined position (fig.27).

This case appears to happen just as the one mentioned above by mere chance. The initial shaft falls down to one side under an angle of roughly 90 degrees compared to the symmetry-plane and leans against the soft body. However, the shells continues to grow in the symmetry-plane. In this case the excess weight appears too large, and the growing shell moves more and more to the other side of the soft body. Again this fact can be derived from the sculpture, now with an opposite rib direction (fig. 27e).



Fig. 27. Ontogenetic stages in the helically coiled heteromorph ammonite Eubostrychoceras. After K. Tanabe et al. (1981), modified


According to fig. 26c the shell must be balanced by muscular pull. This fact becomes manifest by the asymmetrical sculpture which indicates that an increased pull has to be employed near the outer shell margin. The growing shell winds around the initial shaft. In the position as in fig. 26d a balanced weight situation again is present, as demonstrated by the straight ribs. Actually, the shell now might grow again in its symmetry-plane. In other cases this eally happens. However, this Eubostrychoceras is unsuccessful and the growing shell nves now more and more to the other side of the soft body, which requires an opposite muscular pull. This becomes evident in the changed rib direction



Abb. 28. Early ontogenetic stages in the helicoid heteromorph Madagascarites. After K. Tanabe et al. (1981), slightly modified.


Fig.28 presents early growth stages of the heteromorph Madagascarites which shows clearly the transition from the first straight shaft to the helicoid stages. The weight increase in this form apparently was rather low which results in a less tightly coiled helicoid form. There must be serious doubts whether all these heteromorphs represent different biospecies.

 Increasing muscular pull leads to a growing sutural complication.



  The initial shaft topples abruptly over and assumes a position inclined under 45°  

                         with respect to the symmetry-plane (fig. 29)




Fig.29. Ontogenetic stages in corkscrew-like coiled heteromorph ammonites, Hyphantoceras or Bostrychoceras


This is the third and last possibility of shell form which can be built if the initial shaft abruptly topples over following a strong weight increase. Since now the initial shaft can no more be touched by the growing shell a corkscrew-like form develops. Finally, it becomes so large that it hangs down from the soft body. By chance, it can hang down in the symmetry-plane. This can coincide with the termination of shell growth. However, in other cases the further growth happens as a normal spiral as in Heteroceras. Since the angle between the oblique initial shaft can vary between 0° and 90°, there are many possibilities of variation between open corkscrew-like forms and tightly helically coiled shells such as Turrilites.

           The hank-like Nipponites

Finally, in fig.30 ontogenetic stages of Nipponites are shown. I must admit that I had the greatest problems with this form for a long time, mainly because I had never seen the the initial whorl. Only the idealized shape presented by T. Okamoto (1988) made clear to me how this ammonite could grow. By the way, on occasion of a cephalopod symposium in Lyon in 1990 I had the opportunity to meet T. Okamoto and his mentor K. Tanabe.



Fig.29. Ontogenetic stages in Nipponites, as well as a cut through the initial plane of an idealized specimen, in part after Okamoto.


In this form the open initial shell moves completely out of the symmetry-plane, presumably due to a hampering influence by the soft body. The transversely positioned initial shell prevents the formation of a normally coiled shell. The growing shell has to give way to the initial whorl. Consequently it rolls during growth from one side to the other, according to the position of the shell’s centre of gravity.

It should be noted that the formation and individual development in all of these forms which so far appeared entirely incomprehensible becomes only possible by the assumption of a bottom-related lifestyle. Surely, the growth of normally coiled shells can well be explained in this way. However, an appropriate model must be suitable to explain any form. But the heteromorph ammonites have always been the most difficult and mysterious forms.

I hope I could make understandable that the former idea of floating or swimming ammonites does not hold true and that these animals without any exception have been bottom-dwellers.