DISCUSSION
Our understanding of how organisms respond to predation risk has traditionally focused on a small number of specific traits in only a few environments. Recent advances have moved the standard of plasticity research to a multi-trait approach including morphology, life history and behaviour along environmental gradients. This has been complemented by a growing appreciation that morphometric analyses applied to organism shape can provide added value to analyses of phenotypic plasticity.
Here, building on this growing use of morphometrics in plasticity research, we evaluated ‘shape’ plasticity among three genotypes ofD. pulex exposed to a gradient of six levels of predation risk. Our objective was to use morphometric shape as a ‘summary’ trait affected by responses to predation risk in life history and morphology to evaluate several hypotheses about the impact of predation risk on whole organism plasticity. Our motivation was linked to size selective predation theory and work on the response of fish to predation risk where size and shape are both linked to survival . We found that theD. pulex response to predation risk involves both modular and integrated changes, with changes in the inducible neckteeth defence linked to changes in head and body shape. This suggests that there is a complex response to predation, including strong developmental correlations in how Daphnia body plans are organised.
Our first set of results from the trajectory analysis showed how morphology changed in response to a gradient of predation risk. In accordance with previous studies of the inducible defence and our initial hypothesis, we found that head width (neck-change) increases along the gradient of predation risk . Also, we found that head height varied among clones and body width (belly-bulge) was a non-linear feature of change. This last result is contrary to our hypothesis derived from the fish literature; we predicted that body width would decrease along a rising gradient of predation risk linked to swimming/evasion, but in reality, body width varies regardless of the level of predation risk.
Our second set of results from the modularity and integration tests showed evidence for modularity between the head and the body, as expected, but also integration across these two body regions. This suggests that there is a high level of coordination within the head and the body, but there is also an integrated ‘trade-off’ among aspects of the body plan under predation risk. Specifically, as we predicted, as the neck region enlarges, the back moves higher, the tail is drawn towards the front of the body and the snout shifts backwards. Contrary to our other hypothesis, we found no evidence for modularity in the back and front regions. However these two groupings of landmarks are integrated in their response to predation risk, which is likely to be derived from the modular and integrated response of the head and lower body.
It is well-documented that the inducible defence of D. pulexincreases in response to higher levels of predation risk . This morphological defence has long been hypothesized to form part of a whole-organism response to predation that includes changes in size and life-history traits linked to the allocation of energy to growth, defence morphology and reproduction .
In comparison, there has been some work on body width, which is another key aspect of shape that we analysed, but it is unclear what role this plays in the response to predation risk. Previous studies have used linear morphometrics to show a relatively small increase in body width in defended compared to undefended morphs of D. pulex andD. magna which suggests that body width may play a minor role in the response to predation. In support of this, it has been shown that body width is a better predictor for prey size range in Chaoborusthan body length . Chaoborus usually swallows prey that cannot be deformed only if its diameter is not wider than the larva’s head capsule diameter . The increase in body depth might additionally make prey handling for the larvae more difficult. Alternatively, the increased body width in defended compared to undefended morphs may result from the increase in strength, and possibly thickness, of the carapace . Similar to the changes in body width, variation in head height, something revealed in our analysis, has not been well-documented, but may be linked to hydrodynamics .
In this study, there was both a modular and whole-organism, integrated response to predation (sensu . This is entirely possible, as modularity and integration are not two ends of a continuum, but represent two mutually compatible destinations of selection. It is commonly thought that the evolution of modularity and integration is linked to trait functionality. In the case of modularity, there can be selection for ‘variational adaptation’, where traits that often respond together to environmental pressures, such as predation risk, are integrated into one module, and traits that rarely need to be changed at the same time are packed into another module . This may explain why changes in the head of D. pulex , which form the main response to predation risk, are relatively independent of changes in the body.
Possible functions of the observed integration could be for carapace stability or an improved escape response . Alternatively, integration could increase fitness by co-ordinating the response of multiple traits to spatial and/or temporal variation in the environment. Regardless, our results deliver strong support for the existence of developmental constraints (correlations) that are worth investigating to understand the nature of predator induced phenotypic plasticity.
Shape variation in response to predation risk has not previously been studied in D. pulex . In this study, we exposed D. pulex to six levels of predation risk and evaluated shape plasticity using geometric morphometrics and phenoptypic trajectory analysis. We now have a better understanding that D. pulex shape is a multivariate response to predation risk, that there is genetic variation in this response and that the responses can be both modular and integrated, with associated adaptive and non-adaptive (constraint) hypotheses in need of further evaluation. Thus, there are two clear ‘next-steps’. The first is to establish the adaptive benefit/costs of shape change and variation. The second is the molecular ecology of the developmental constraints. The availability of genomic tools, the clonal nature of Daphnia , well-established experimental protocols and recent high throughput image analysis are an outstanding platform for future research.