According to J.S. Huxley (1949) individuality is relative and never complete or perfect; there are degrees of individuality which can be attributed to different levels from cells to colonies: ‘Many animals such as polyps form colonies whose members are attached to a common living stem. All gradations occur between single polyps, those forming temporary colonies by budding, and those with permanent colonies. In permanent colonies, all members may be alike, or there may be division of labour among different types. In some extreme cases, the colony behaves as a single unit …, the members being entirely subordinate.’ (Huxley, 1949, p. 256). The relativity of individuality has been repeatedly noted during the last fifty years, but the question has only recently gained a broader interest among evolutionary biologists, as reviewed by Pineda-Krch & Lehtilä (2004).
The adoption of the genic view of adaptation (Williams, 1966) was an important step towards this direction as individual organisms were set aside of the genes – the proposed ultimate units of selections. Still, most attributes of the old ‘Weismannian’ concept of an individual were associated with the ‘gene vehicles’ which should show a sufficient degree of integration, their parts should share the same genes through out, and the parts should share of a single, common exit channel into the future (Dawkins, 1989). Genetically heterogeneous plants would not qualify as true gene vehicles (Dawkins, 1982). Recent approaches to the evolution of individuality have also been mostly devoted to the evolution of the germ line (Buss, 1987). By sequestering a germ line from somatic cells early in development, the opportunity for genetic variation and evolutionary change within an organism will be limited, and evolution will to a greater extent depend on the covariance of organism-level fitness and zygote genotype (Michod, 1999).
In Fig. 1, the evolution of individuality is, for simplicity, perceived in two steps from a group of lower-level units which are neither integrated nor specialized. In the first step, the lower-level units become physically and/or functionally integrated into a more inclusive unit. In the next step, they become specialized to show division of labour as Huxley (1949) described in the case of hydrozoan colonies. The second step may not proceed very far if genetic changes within an organism are not restricted. Dawkins (1989), p. 264) considered that ‘the essential, defining feature of an individual organism is that it is a unit that begins and ends with a single-celled bottleneck’. A single-celled bottleneck ensures, to some degree at least, genetic homogeneity over the developing body (Pineda-Krch & Lehtilä, 2004: paths 7–10 in Fig. 1) and, what is even more important, it enables a regulated development and specialization of the component cells into different functions within the same body (Dawkins, 1989). The sequestration of the germ line can lead in the same direction (Buss, 1987; Michod, 1999).
From this perspective, intraorganismic genetic heterogeneity (IGH) appears as a cost rather than a benefit if high IGH constrains evolution towards increasing specialization among component units (Fig. 1b). However, might there be some benefits of IGH in the first step when the more inclusive unit is built up by integration of more or less unspecialized lower-level units (Fig. 1a)? Although Pineda-Krch & Lehtilä (2004) list a number of potential benefits of IGH, only a few studies have so far documented such benefits. Apart from possible size-related advantages, which may have been important in the early evolution of multicellularity (Bonner, 2000), the benefits of IGH may be rare. In the case of long-lived plants, IGH might provide a potential way to defend against short-lived pests, as Pineda-Krch & Lehtilä (2004) have discussed, but this may not be a predominant tactic to cope with temporal and spatial heterogeneity. Most plants are characterized by a high degree of phenotypic plasticity in response to environmental heterogeneity (Grime & Mackey, 2002), which is, for example, reflected in relatively high within-tree variation in herbivore resistance (Suomela & Ayers, 1994).
I complete agree with Pineda-Krch & Lehtilä (2004) that we should broaden our views beyond the ‘Weismannian’ individuality, and their approach provides interesting challenges for further theoretical and empirical studies on IGH. With regards to the evolution of individuality, one may further question whether the restriction of IGH really is the defining issue in the transition from a lower to a higher level.
Genetic homogeneity as such is not sufficient to awaken the higher-level unit into a life of its own. What is further needed is the emergence of genetically defined phenotypes for the higher-level units. Moreover, the phenotypic traits at the higher level should have fitness consequences at the levels where reproduction and propagation of genes to future generations take place. When these higher-level fitness effects are strong enough, they may drive evolution towards increased integration and specialization among the lower-level units to a degree that they appear more or less subordinate to vital functions of the more inclusive unit. One can only guess the selection forces which may have sped up specialization among cells in a multicellular colony: selection for increased mobility or, in the case of sessile colonies, for vertical growth requiring structural stability may well be among them (e.g. Bonner, 2000). In any case, the benefits of specialization gained by fertile colony-members must have been remarkable in order to compensate for the lost reproductive potential of those colony-members from which an opportunity for reproduction of their own has been omitted.
Consequently, the costs and benefits of integration and specialization among component units may be equally or even more essential in the evolution of individuality than those of intraorganismic genetic variation.