Selective Regimes Shaping Morphologies

Last month, I wrote on the Dendrosenecios and their diversification throughout the tropical mountains of eastern and central Africa.  The Dendrosenecios were incredibly well adapted to their harsh environment.  The remarkable features that they show, such as: the large pith volume, the marcescent foliage and nyctinasty were adapted to their equatorial, tropical alpine environment.  Interestingly, a few other places around the world, such as Hawaii and the tropical Andes have similar climates.  In response, other plants within the family Asteraceae have developed the same adaptations as the Dendrosenecios of Africa.  This is known as convergent evolution, which can be simply defined as similar adaptations to similar environments by members of distantly related lineages.

Figure 1.  Asteraceae convergent evolution: The Silversword Alliance of Hawaii (left) and the Dendrosenecios of Africa (right).

This month, I will look at a local example of convergent evolution that is relevant to my Honours Research Project, focusing on scatter-hoarding in the Proteaceae.  There are three major guilds of seed dispersal in the fynbos, two of them are serotiny and myrmechocory.  Serotinous seeds are typically held in the plant canopy and dispersed by wind, post-fire.  Myrmechocorous seeds have a nutritious tissue called an elaiosome attached to the seed hull.  Ants drag these seeds underground and consume the elaiosome, leaving the seed intact and protected from seed predators and fires.  The third, scatter-hoarding, is a less understood dispersal guild in the fynbos.  Seeds of this guild are generally large with thick hulls, lacking a wing or elaiosome.

Figure 2.  A typical scatter hoarding seed.  This is an image of the tooth marks of Rhabdomys pumilio, known as the number one seed predator, on the hull of a Ceratocaryum argenteum (Restionaceae) seed.

Scatter-hoarding takes place when rodents are presented with a glut of seeds and do not have the appetite to eat them all.  In certain species, a behavioral response is then triggered – they bury the remaining seeds, typically away from where the seeds were found.  They will then rely on their smell and memory to relocate the seeds they have buried, extending their use of the food source.  Of course, not all of the seeds are relocated, allowing a dispersal advantage to the lucky seeds that got away, as they will now not need to compete with their mother plant. 

So, within the broader scope of scatter-hoarding, are a number of interesting questions that can be asked about this behavior.  A recent area of research has been the selection for specific seed traits, such as hull thickness, by predating and scatter-hoarding rodents.  Rusch et al (2012) expected that rodents would preferentially disperse and cache large, thick hulled seeds but consume small, thin hulled seeds as they are found. 

Initially, the natural variation of the two seed traits was quantified by extensive measurements of all parameters for Leucadendron sessile seeds – the suncone bush.  The research team then used an interesting technique to better understand seed selection based on hull thickness and seed size – because it is rather difficult to determine the thickness of the hull or fleshiness without breaking the hull open.  Firstly, the seed hulls of L. sessile seeds were removed and replaced with a non-toxic, plumber’s putty as a simulant of a woody hull.  This way, the thickness of the seed hull was controlled for.  The putty was placed on the naked seeds at 1, 2 or 2.5mm.  Next, macadamia nuts were used as a replacement for L. sessile seeds, because their size could be controlled for.  The macadamias were sanded down to either be at the recorded smallest, average or largest sizes and then covered in a standard 2mm hull thickness.  Seeds were then placed out overnight at seed stations at the research site.

Figure 3. Acomys subspinosus - the Cape spiny mouse. 

The biggest culprit of seed removal was suggested to be Acomys subspinosus or better known as the Cape Spiny Mouse.  The results of this study are shown in the two figures below.  Figure 4 shows the fate of seeds in the hull thickness experiment.  Figure 5 shows the fate of seeds in the seed size experiment.  Interpretation of the results can be seen in the figure captions.

Figure 4.  Fates of thin-hulled, average-hulled and thick-hulled L. sessile seeds after a 12 hour exposure to Cape Spiny mice in the field.  Thin-hulled seeds were eaten most often, average-hulled seeds cached and thick-hulled seeds were not dispersed. 

Figure 5.  The fates of small, average and large macadamia seeds after exposure to the Cape Spiny mice for 12 hours in the field.  Small seed were predominantly eaten, average sized seeds cached and large seeds not dispersed.

It is clear from this research that rodents have a preference for which seeds they chose to disperse or consume.  Average sized seeds with an average size and hull-thickness were preferentially dispersed.  While seeds that were small or thin-hulled were most often consumed at the seed stations.  Large or thick-hulled seeds were frequently left at the depot sites.  The authors of this research suggest that the energetic cost of handling, transport and burial are the reasons for their preferential choices.  The importance of these results is that it suggests that rodents exert stabilizing selection against the extremes of seed size and hull thickness for L. sessile seeds.

In the extreme environments of the tropical alpine zone, the climate creates a selection regime that leads to the distinct morphological features that are shown by members of the Asteraceae.  Around my home, Cape Town, fynbos plants such as those members of the Proteaceae and Restionaceae are exerted to a selective regime by seed consuming and dispersing rodents.  Seed predators are common throughout the fynbos and therefore it has been suggested that the scatter-hoarding guild of seed dispersal is more common than is currently understood.  Very few of the relationships between plants and their scatter-hoarding rodents have been documented, which allows for exciting future research for a young ecologist like me. 

(Rusch, U.D., Midgley, J.J., Anderson, B.  2013.  Rodent consumption and caching behaviour selects for specific seed traits. South African Journal of Botany.  84: 83-87)