The forming of a thunderegg

The forming of a thunderegg.

A lithophyse, also known as thunderegg, is a more or less round stone. The translation of the Greek word lithophysa is ‘ stone cavity ‘. The conception was introduced in 1860 by the Hungarian Von Richthofen. For collectors, it is interesting when the cavity is filled in with crystals (so-called geode) or/and with agate, opal and jasper. But whether there is a cavity in that round shaped stone, and if this demanded filling really is there, is not always fixed.

When the cavity is not present, that round stone is called a spherulite or more popular: a mudball or a dud.


Spherulites are found in silica rich volcanic deposits (ignimbrite, rhyolite). Research has shown that spherulites commonly are found in a transition zone between slowly cooled, crystallized deposits and deposits in which crystallization, as the result of rapid cooling down, was not possible (perlite, volcanic glass). The volcanic deposits can be of magmatic (rhyolite) aswell pyroclastic (ashflow; ignimbrite) origin. How a spherulite exactly forms is still being discussed.

Several researchers assume that a spherulite forms in a more or less liquid, viscous lava flow, and whose growth yet for the solidifying of the flow stops. The forming of a spherulite in a lava flow starts around a point of nucleation, which can either be a piece of embedded stone, a gasbubble, or a existing biotite, quartz-, or plagioclase crystal. Around the nucleationpoint a zone can be present, which stands out by a different color, hardness and structure: the spherule. Plausible seems to be the theory of Robert Paul Colburn, a man who literally has explored, dug, studied and cut thundereggs lifelong. He has written some notable publications. According to Colburn a spherulite (which often shows a radial patterned crystallization), consists of high-cristobalite (a polymorph of silicium dioxide). During the formation of cristobalite crystals around the nucleation point, water vapour is set free. The occurance of the watervapour makes diffusion of silica (building block of cristobalite) towards the growing cristobalite crystals possible. This process, once set in motion, seems to strengthen itself and continues until one or more limiting conditions (right temperature, presence of silica and water) fades.

Measurements have shown that  spherulites from the Rockhound State Park (New Mexico, USA) exist of 81% cristobalite. The perlite in which the spherulites were found, only held 1.5% cristobalite. Whereas the chemical composition of spherulite and perlite were about the same. Moreover, the texture of the spherulites cristobalitecrystals provide, generally spherulites are less weathered than the hostrock in which they occur.

To find out more about the subject and see examples of nucleation points and spherules, visit the core corner.

A cut spherulite from Cubuk, Turkey, 7 cm. In the centre a spherule is visible. The radial pattern of cristallization is recognizable due to weathering.

A cut spherulite from Cubuk, Turkey, 7 cm.
In the centre a spherule is visible. The radial pattern of crystallization in the shell is recognizable due to weathering.

Lithophyses (thundereggs).

As the name indicates, a cavity exists in a lithophyse. Concerning the forming of the cavity, there is a theme of discussion, again …

Simplyfied figures based on Trunk Diagrams (Colburn, 2003, 2004).

Fig. 1. Simplified figures based on Trunk Diagrams (Colburn, 2003, 2004).

Some researchers assume that the thunderegg and the gasbubble within it, occured at the same time. Denting, cooling and/or pressure relief would determine the final shape of the cavity in the still plastic lithophyse.

Other researchers think that a spherulite is at the base of the formation of a lithophyse (fig.1 A). Further crystallization of cristobalite takes place on a spherulite, while at the same time, the opening of a cavity occurs (fig.1 B). The formation of a cavity could be the result of pressure change and volume reduction in the lava and spherulite during the cooling process. In many thundereggs the growth of the cristobalite crystals were interrupted; the thunderegg grew in stages. Every stage is marked by, what Paul Colburn named, Fronts Of Spherulitic Crystallization (FOSC). Often these are not visible with bare eye,  a (electro)microscope is needed to notice them.

According to Colburn the origin and the size of the cavity is the result of internal gas pressure. The excess pressure in the cavity is caused by the release of water vapor as result of crystallization of cristobalite. Heat emission as a result of the crystallisation process can increase the excess pressure even further. In addition, the volume loss associated with this crystallization and the weight of any existing deposits on top of the thunderegg lead to final shape and size of the cavity.

Further crystallization of cristobalite keeps the gas within the lithophyse; growth of the thunderegg and further opening of the cavity occur simultaneously (fig.1 C).

At the location where the ends of the starpoints of a cavity reaches the outside of the lithophyse, sometimes an extra, sharp ridge has been formed, that prevents to escape the presure from the thunderegg. These are called pressure ridges (fig.1 D).

Moreover, at some jasper filled thundereggs, it looks like that there has been a vacuum in the cavity at first. Material coming from outside, so to speak, possibly was sucked inward the thunderegg.



Overlay drawings on a Mirador thunderegg, based on the idealized grow model by Bryan (1934).

Fig. 2. Overlay drawings on a Mirador thunderegg, based on the idealized growth model by Bryan (1934).

Radial growpatterns and the cavity.
Reading about the forming of a thunderegg I came across a drawing of the idealized growth model by Bryan (1934). It shows how the radial growth of the cristobalite crystals in the shell of a thunderegg keep up with the opening of a cavity. Some lithophyses do show the drawn structures: I made an overlay drawing on a thunderegg from Mirador / Argentina (fig.2 A), which shows the patterns pretty well.

In Mirador thundereggs the cristobalite crystalbundles are visible well. An area with a distinct structure is marked with a green line in the figures: the spherule. In the centre of the spherule lies the nucleation point. The thundereggs shell around the spherule can be subdivided in two sections. The blue lined sections have cristobalite bundles that point towards the nucleation point of the egg, at least when you pretend if the red sections and the cavity weren’t there; the blue lined parts do fit nicely!

The shell in the red sections has been growing under a different angle. The starpoints opening during the growth of the egg seem to have caused in some way a reaction of new starting cristobalite crystalbundles. In this way the shell could grow along with the cavity and the cavity was permanently concluded from the adjacent occuring conditions.

Thinking backward in the forming of the egg in Figure 2, C is showing how the starting of the growth of the egg must have been and it matches the situation as given by Paul Colburn in his trunkdiagram (see ‘based on’ figure 1 B above). Figure 2 C is a compiled (cut and paste) figure from figure 2 B alongside.

The growing thunderegg in 2 C has barely a cavity, but it is opening up while the cristobalite bundles are growing outwards; they just are following their initial radial pattern outwards.

The radial patterns aren’t visible (well) in thundereggs from many locations, but the eggs from Mirador are outstanding showing them. Microscopic views of eggs from various sites often do confirm the existing of radial patterns.


More interesting facts about the components of a thunderegg: the shell, cavity and filling.